Skip to contentSkip to conversation

automount won't start - just hangs

DISCUSSION Jim Duda 2 participants 17 messages Dec 31, 2007 – Jan 7, 2008 Kernel automounter

lore
Share & tools
lore

Jim Duda

author
#0 lore raw

I'm attempting to use autofs on some diskless machines which operate using an NFS mounted root file system. I have three different computers which share the same root file system (with unique /tmp and /var directories) and the same 2.6.22 kernel.

I want to use autofs for removable USB memory sticks.

On one machine, I can use autofs just fine. Autofs mounts and unmounts the usb sticks as I expect it to. Works nice.

However, on the other two machines, I cannot get automount to start. It simply hangs upon startup, I cannot ^C to exit or anything. I have to kill -9 the process.

I cannot figure out what automount is looking for, and not getting, which would cause it to hang. I've attempted to use gdb to help me find the problem, but I haven't had much luck. (I've downloaded the most recent 5.0.2 tar bar and build automount with -g to debug.

I'm hoping someone might shed some light on how best to debug this issue.

On the two machines which fail, I see this message appear on the NFS server which is serving up the root file system.

Dec 31 16:05:50 linux mountd[2130]: authenticated unmount request from 

asterisk:621 for /tftpboot/muekow/i386 (/tftpboot/muekow)

I don't see any such message on the machine which works properly (the only clue I currently have).

On the diskless machines, I have these mount points already defined before I attempt to start automount.

192.168.1.254:/tftpboot/muekow/i386 on /nfsroot type nfs (rw,vers=3,rsize=32768,wsize=32768,hard,nolock,proto=udp,timeo=7,retrans=3,sec=sys,addr=192.168.1.254) none on /proc type proc (rw)
none on /dev type tmpfs (rw)
/sys on /sys type sysfs (rw)
none on /proc/bus/usb type usbfs (rw)
192.168.1.254:/home on /home type nfs (rw,vers=3,rsize=32768,wsize=32768,hard,nolock,proto=udp,timeo=7,retrans=3,sec=sys,addr=192.168.1.254) 192.168.1.254:/media on /media type nfs (rw,vers=3,rsize=32768,wsize=32768,hard,nolock,proto=udp,timeo=7,retrans=3,sec=sys,addr=192.168.1.254)

Does automount for any reason attempt to unmount nfs mount points which are already in place?

Thanks for any advice, I'm really stumped as to why automount won't start.

Jim

maintainer
#1 lore raw
Quoted 10 lines · Jim Duda

I'm attempting to use autofs on some diskless machines which operate using an NFS mounted root file system. I have three different computers which share the same root file system (with unique /tmp and /var directories) and the same 2.6.22 kernel.

I want to use autofs for removable USB memory sticks.

On one machine, I can use autofs just fine. Autofs mounts and unmounts the usb sticks as I expect it to. Works nice.

Quoted 8 lines

However, on the other two machines, I cannot get automount to start. It simply hangs upon startup, I cannot ^C to exit or anything. I have to kill -9 the process.

I cannot figure out what automount is looking for, and not getting, which would cause it to hang. I've attempted to use gdb to help me find the problem, but I haven't had much luck. (I've downloaded the most recent 5.0.2 tar bar and build automount with -g to debug.

Did you apply any of the patches for 5.0.2?

Quoted 2 lines

I'm hoping someone might shed some light on how best to debug this issue.

Have a look at http://people.redhat.com/jmoyer for what to collect and post here.

Ian

author
#2 lore raw

I have gathered the information required in this post. I have provided the common information for machines which work and fail. I have provided the syslog debug message for those that work and also those that fail.

lroom# rpm -q autofs
autofs-5.0.1-31

lroom# uname -a
Linux lroom 2.6.22.2-tftp #1 Sun Aug 12 11:14:47 EDT 2007 i686 i686 i386 GNU/Linux

lroom# cat /etc/auto.master
#
# Sample auto.master file
# This is an automounter map and it has the following format

# key [ -mount-options-separated-by-comma ] location
# For details of the format look at autofs(5).
#
#/misc  /etc/auto.misc
#/net   -hosts
/mnt/usb /etc/auto.usb --timeout=2
#
# Include central master map if it can be found using
# nsswitch sources.
#
# Note that if there are entries for /net or /misc (as
# above) in the included master map any keys that are the
# same will not be seen as the first read key seen takes
# precedence.
#
#+auto.master

lroom# cat /etc/auto.usb

device_a         -fstype=vfat,rw,gid=100,umask=002       :/dev/sda1
device_b         -fstype=vfat,rw,gid=100,umask=002       :/dev/sdb1

lroom# cat /etc/nsswitch.conf
#

# /etc/nsswitch.conf
#
# An example Name Service Switch config file. This file should be
# sorted with the most-used services at the beginning.
#
# The entry '[NOTFOUND=return]' means that the search for an
# entry should stop if the search in the previous entry turned
# up nothing. Note that if the search failed due to some other reason
# (like no NIS server responding) then the search continues with the
# next entry.
#
# Legal entries are:
#
#       nisplus or nis+         Use NIS+ (NIS version 3)
#       nis or yp               Use NIS (NIS version 2), also called YP
#       dns                     Use DNS (Domain Name Service)
#       files                   Use the local files
#       db                      Use the local database (.db) files
#       compat                  Use NIS on compat mode
#       hesiod                  Use Hesiod for user lookups
#       [NOTFOUND=return]       Stop searching if not found so far
#

# To use db, put the "db" in front of "files" for entries you want to be

# looked up first in the databases
#
# Example:
#passwd:    db files nisplus nis
#shadow:    db files nisplus nis
#group:     db files nisplus nis
passwd:     files
shadow:     files
group:      files

#hosts: db files nisplus nis dns
hosts: files dns

# Example - obey only what nisplus tells us...

#services:   nisplus [NOTFOUND=return] files
#networks:   nisplus [NOTFOUND=return] files
#protocols:  nisplus [NOTFOUND=return] files
#rpc:        nisplus [NOTFOUND=return] files
#ethers:     nisplus [NOTFOUND=return] files
#netmasks:   nisplus [NOTFOUND=return] files

bootparams: nisplus [NOTFOUND=return] files

ethers:     files
netmasks:   files
networks:   files
protocols:  files
rpc:        files
services:   files

netgroup: files

publickey: nisplus

automount:  files
aliases:    files nisplus

lroom# cat /etc/sysconfig/autofs
#
# Define default options for autofs.
#
# MASTER_MAP_NAME - default map name for the master map. #
#MASTER_MAP_NAME="auto.master"
#
# TIMEOUT - set the default mount timeout (default 600). #
TIMEOUT=300
#
# BROWSE_MODE - maps are browsable by default.
#
BROWSE_MODE="no"
#
# APPEND_OPTIONS - append to global options instead of replace. #
#APPEND_OPTIONS="yes"
#
# LOGGING - set default log level "none", "verbose" or "debug" #
LOGGING="debug"
#
# Define the default LDAP schema to use for lookups
#
# System default
#
#MAP_OBJECT_CLASS="nisMap" #ENTRY_OBJECT_CLASS="nisObject" #MAP_ATTRIBUTE="nisMapName"
#ENTRY_ATTRIBUTE="cn"
#VALUE_ATTRIBUTE="nisMapEntry"
#
# Other common LDAP nameing
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount"
#MAP_ATTRIBUTE="ou"
#ENTRY_ATTRIBUTE="cn" #VALUE_ATTRIBUTE="automountInformation"
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount" #MAP_ATTRIBUTE="automountMapName" #ENTRY_ATTRIBUTE="automountKey" #VALUE_ATTRIBUTE="automountInformation"
#
# AUTH_CONF_FILE - set the default location for the SASL
# authentication configuration file.
#
#AUTH_CONF_FILE="/etc/autofs_ldap_auth.conf"
#
# General global options
#
#OPTIONS=""
#

The "linux" machine is the NFS server which provides the root file system for the "lroom" and "gym" machines. The /tftpboot/muekow/i386 directory is the root file system for the diskless machines, which is a static hard mount setup during the early boot process.

Here are the log files for the "lroom" machine which works properly.

Jan  1 11:43:37 lroom automount[1968]: Starting automounter version 
5.0.1-31, master map auto.master
Jan  1 11:43:37 lroom automount[1968]: using kernel protocol version 5.00
Jan  1 11:43:37 linux mountd[2130]: authenticated unmount request from 
lroom:915 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 linux mountd[2130]: authenticated unmount request from 
lroom:637 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 lroom automount[1968]: mounted indirect mount on 
/mnt/usb with timeout 2, freq 1 seconds
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry 
/mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry 
/mnt/usb/dcim
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/dcim
Jan  1 11:45:17 lroom xinetd[1820]: START: telnet pid=2280 

from=192.168.1.254

Here are the log files for the "gym" machine which fails to start automount properly. The automount program is in the hung state at this point.

Jan  1 11:49:36 gym automount[1846]: Starting automounter version 
5.0.1-31, master map auto.master
Jan  1 11:49:36 gym automount[1846]: using kernel protocol version 5.00
Jan  1 11:49:36 linux mountd[2130]: authenticated unmount request from 

gym:778 for /tftpboot/muekow/i386 (/tftpboot/muekow)

I realized today that the machines which fail to work are P3 processors whereas the one that works is a P4 processor. I don't know if that has meaning or is a pure coincidence.

Thanks for the help.

Jim

Quoted 28 lines · Ian Kent
Quoted 10 lines · Jim Duda

I'm attempting to use autofs on some diskless machines which operate using an NFS mounted root file system. I have three different computers which share the same root file system (with unique /tmp and /var directories) and the same 2.6.22 kernel.

I want to use autofs for removable USB memory sticks.

On one machine, I can use autofs just fine. Autofs mounts and unmounts the usb sticks as I expect it to. Works nice.

Quoted 8 lines

However, on the other two machines, I cannot get automount to start. It simply hangs upon startup, I cannot ^C to exit or anything. I have to kill -9 the process.

I cannot figure out what automount is looking for, and not getting, which would cause it to hang. I've attempted to use gdb to help me find the problem, but I haven't had much luck. (I've downloaded the most recent 5.0.2 tar bar and build automount with -g to debug.

Did you apply any of the patches for 5.0.2?

Quoted 1 line

I'm hoping someone might shed some light on how best to debug this issue.

Have a look at http://people.redhat.com/jmoyer for what to collect and post here.

Ian

maintainer
#3 lore raw
Quoted 4 lines · Jim Duda

I have gathered the information required in this post. I have provided the common information for machines which work and fail. I have provided the syslog debug message for those that work and also those that fail.

This all looks OK but the debug log doesn't have the debugging info. Did you set syslog to send daemon.* to the log?

I'm not sure we'll get much more from it but we should at least have a look at it.

Do you get any core files for automount?

Quoted 238 lines

lroom# rpm -q autofs
autofs-5.0.1-31

lroom# uname -a
Linux lroom 2.6.22.2-tftp #1 Sun Aug 12 11:14:47 EDT 2007 i686 i686 i386 GNU/Linux

lroom# cat /etc/auto.master
#
# Sample auto.master file
# This is an automounter map and it has the following format

# key [ -mount-options-separated-by-comma ] location
# For details of the format look at autofs(5).
#
#/misc  /etc/auto.misc
#/net   -hosts
/mnt/usb /etc/auto.usb --timeout=2
#
# Include central master map if it can be found using
# nsswitch sources.
#
# Note that if there are entries for /net or /misc (as
# above) in the included master map any keys that are the
# same will not be seen as the first read key seen takes
# precedence.
#
#+auto.master

lroom# cat /etc/auto.usb

device_a         -fstype=vfat,rw,gid=100,umask=002       :/dev/sda1
device_b         -fstype=vfat,rw,gid=100,umask=002       :/dev/sdb1

lroom# cat /etc/nsswitch.conf
#

# /etc/nsswitch.conf
#
# An example Name Service Switch config file. This file should be
# sorted with the most-used services at the beginning.
#
# The entry '[NOTFOUND=return]' means that the search for an
# entry should stop if the search in the previous entry turned
# up nothing. Note that if the search failed due to some other reason
# (like no NIS server responding) then the search continues with the
# next entry.
#
# Legal entries are:
#
#       nisplus or nis+         Use NIS+ (NIS version 3)
#       nis or yp               Use NIS (NIS version 2), also called YP
#       dns                     Use DNS (Domain Name Service)
#       files                   Use the local files
#       db                      Use the local database (.db) files
#       compat                  Use NIS on compat mode
#       hesiod                  Use Hesiod for user lookups
#       [NOTFOUND=return]       Stop searching if not found so far
#

# To use db, put the "db" in front of "files" for entries you want to be

# looked up first in the databases
#
# Example:
#passwd:    db files nisplus nis
#shadow:    db files nisplus nis
#group:     db files nisplus nis
passwd:     files
shadow:     files
group:      files

#hosts: db files nisplus nis dns
hosts: files dns

# Example - obey only what nisplus tells us...

#services:   nisplus [NOTFOUND=return] files
#networks:   nisplus [NOTFOUND=return] files
#protocols:  nisplus [NOTFOUND=return] files
#rpc:        nisplus [NOTFOUND=return] files
#ethers:     nisplus [NOTFOUND=return] files
#netmasks:   nisplus [NOTFOUND=return] files

bootparams: nisplus [NOTFOUND=return] files

ethers:     files
netmasks:   files
networks:   files
protocols:  files
rpc:        files
services:   files

netgroup: files

publickey: nisplus

automount:  files
aliases:    files nisplus

lroom# cat /etc/sysconfig/autofs
#
# Define default options for autofs.
#
# MASTER_MAP_NAME - default map name for the master map. #
#MASTER_MAP_NAME="auto.master"
#
# TIMEOUT - set the default mount timeout (default 600). #
TIMEOUT=300
#
# BROWSE_MODE - maps are browsable by default.
#
BROWSE_MODE="no"
#
# APPEND_OPTIONS - append to global options instead of replace. #
#APPEND_OPTIONS="yes"
#
# LOGGING - set default log level "none", "verbose" or "debug" #
LOGGING="debug"
#
# Define the default LDAP schema to use for lookups
#
# System default
#
#MAP_OBJECT_CLASS="nisMap" #ENTRY_OBJECT_CLASS="nisObject" #MAP_ATTRIBUTE="nisMapName"
#ENTRY_ATTRIBUTE="cn"
#VALUE_ATTRIBUTE="nisMapEntry"
#
# Other common LDAP nameing
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount"
#MAP_ATTRIBUTE="ou"
#ENTRY_ATTRIBUTE="cn" #VALUE_ATTRIBUTE="automountInformation"
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount" #MAP_ATTRIBUTE="automountMapName" #ENTRY_ATTRIBUTE="automountKey" #VALUE_ATTRIBUTE="automountInformation"
#
# AUTH_CONF_FILE - set the default location for the SASL
# authentication configuration file.
#
#AUTH_CONF_FILE="/etc/autofs_ldap_auth.conf"
#
# General global options
#
#OPTIONS=""
#

The "linux" machine is the NFS server which provides the root file system for the "lroom" and "gym" machines. The /tftpboot/muekow/i386 directory is the root file system for the diskless machines, which is a static hard mount setup during the early boot process.

Here are the log files for the "lroom" machine which works properly.

Jan  1 11:43:37 lroom automount[1968]: Starting automounter version 
5.0.1-31, master map auto.master
Jan  1 11:43:37 lroom automount[1968]: using kernel protocol version 5.00
Jan  1 11:43:37 linux mountd[2130]: authenticated unmount request from 
lroom:915 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 linux mountd[2130]: authenticated unmount request from 
lroom:637 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 lroom automount[1968]: mounted indirect mount on 
/mnt/usb with timeout 2, freq 1 seconds
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry 
/mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry 
/mnt/usb/dcim
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/dcim
Jan  1 11:45:17 lroom xinetd[1820]: START: telnet pid=2280 

from=192.168.1.254

Here are the log files for the "gym" machine which fails to start automount properly. The automount program is in the hung state at this point.

Jan  1 11:49:36 gym automount[1846]: Starting automounter version 
5.0.1-31, master map auto.master
Jan  1 11:49:36 gym automount[1846]: using kernel protocol version 5.00
Jan  1 11:49:36 linux mountd[2130]: authenticated unmount request from 

gym:778 for /tftpboot/muekow/i386 (/tftpboot/muekow)

I realized today that the machines which fail to work are P3 processors whereas the one that works is a P4 processor. I don't know if that has meaning or is a pure coincidence.

Thanks for the help.

Jim

Quoted 20 lines · Ian Kent
Quoted 10 lines · Jim Duda

I'm attempting to use autofs on some diskless machines which operate using an NFS mounted root file system. I have three different computers which share the same root file system (with unique /tmp and /var directories) and the same 2.6.22 kernel.

I want to use autofs for removable USB memory sticks.

On one machine, I can use autofs just fine. Autofs mounts and unmounts the usb sticks as I expect it to. Works nice.

Quoted 8 lines

However, on the other two machines, I cannot get automount to start. It simply hangs upon startup, I cannot ^C to exit or anything. I have to kill -9 the process.

I cannot figure out what automount is looking for, and not getting, which would cause it to hang. I've attempted to use gdb to help me find the problem, but I haven't had much luck. (I've downloaded the most recent 5.0.2 tar bar and build automount with -g to debug.

Quoted 8 lines

Did you apply any of the patches for 5.0.2?

Quoted 1 line

I'm hoping someone might shed some light on how best to debug this issue.

Have a look at http://people.redhat.com/jmoyer for what to collect and post here.

Ian

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

author
#4 lore raw

My syslog.conf has this:
*.* @192.168.1.254
Doesn't that cover "everything", including daemon?

No core dumps.

I started using gdb to find where the program was hanging. I was down deep into reading the nisswitch.conf file, then got lost.

Jim

"Ian Kent" <[email protected]> wrote in message news:[email protected]...

Quoted 252 lines · Jim Duda
Quoted 4 lines

I have gathered the information required in this post. I have provided the common information for machines which work and fail. I have provided the syslog debug message for those that work and also those that fail.

This all looks OK but the debug log doesn't have the debugging info. Did you set syslog to send daemon.* to the log?

I'm not sure we'll get much more from it but we should at least have a look at it.

Do you get any core files for automount?

Quoted 238 lines

lroom# rpm -q autofs
autofs-5.0.1-31

lroom# uname -a
Linux lroom 2.6.22.2-tftp #1 Sun Aug 12 11:14:47 EDT 2007 i686 i686 i386 GNU/Linux

lroom# cat /etc/auto.master
#
# Sample auto.master file
# This is an automounter map and it has the following format

# key [ -mount-options-separated-by-comma ] location
# For details of the format look at autofs(5).
#
#/misc  /etc/auto.misc
#/net   -hosts
/mnt/usb /etc/auto.usb --timeout=2
#
# Include central master map if it can be found using
# nsswitch sources.
#
# Note that if there are entries for /net or /misc (as
# above) in the included master map any keys that are the
# same will not be seen as the first read key seen takes
# precedence.
#
#+auto.master

lroom# cat /etc/auto.usb

device_a         -fstype=vfat,rw,gid=100,umask=002       :/dev/sda1
device_b         -fstype=vfat,rw,gid=100,umask=002       :/dev/sdb1

lroom# cat /etc/nsswitch.conf
#

# /etc/nsswitch.conf
#
# An example Name Service Switch config file. This file should be
# sorted with the most-used services at the beginning.
#
# The entry '[NOTFOUND=return]' means that the search for an
# entry should stop if the search in the previous entry turned
# up nothing. Note that if the search failed due to some other reason
# (like no NIS server responding) then the search continues with the
# next entry.
#
# Legal entries are:
#
#       nisplus or nis+         Use NIS+ (NIS version 3)
#       nis or yp               Use NIS (NIS version 2), also called YP
#       dns                     Use DNS (Domain Name Service)
#       files                   Use the local files
#       db                      Use the local database (.db) files
#       compat                  Use NIS on compat mode
#       hesiod                  Use Hesiod for user lookups
#       [NOTFOUND=return]       Stop searching if not found so far
#

# To use db, put the "db" in front of "files" for entries you want to be

# looked up first in the databases
#
# Example:
#passwd:    db files nisplus nis
#shadow:    db files nisplus nis
#group:     db files nisplus nis
passwd:     files
shadow:     files
group:      files

#hosts: db files nisplus nis dns
hosts: files dns

# Example - obey only what nisplus tells us...

#services:   nisplus [NOTFOUND=return] files
#networks:   nisplus [NOTFOUND=return] files
#protocols:  nisplus [NOTFOUND=return] files
#rpc:        nisplus [NOTFOUND=return] files
#ethers:     nisplus [NOTFOUND=return] files
#netmasks:   nisplus [NOTFOUND=return] files

bootparams: nisplus [NOTFOUND=return] files

ethers:     files
netmasks:   files
networks:   files
protocols:  files
rpc:        files
services:   files

netgroup: files

publickey: nisplus

automount:  files
aliases:    files nisplus

lroom# cat /etc/sysconfig/autofs
#
# Define default options for autofs.
#
# MASTER_MAP_NAME - default map name for the master map. #
#MASTER_MAP_NAME="auto.master"
#
# TIMEOUT - set the default mount timeout (default 600). #
TIMEOUT=300
#
# BROWSE_MODE - maps are browsable by default.
#
BROWSE_MODE="no"
#
# APPEND_OPTIONS - append to global options instead of replace. #
#APPEND_OPTIONS="yes"
#
# LOGGING - set default log level "none", "verbose" or "debug" #
LOGGING="debug"
#
# Define the default LDAP schema to use for lookups
#
# System default
#
#MAP_OBJECT_CLASS="nisMap" #ENTRY_OBJECT_CLASS="nisObject" #MAP_ATTRIBUTE="nisMapName"
#ENTRY_ATTRIBUTE="cn"
#VALUE_ATTRIBUTE="nisMapEntry"
#
# Other common LDAP nameing
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount"
#MAP_ATTRIBUTE="ou"
#ENTRY_ATTRIBUTE="cn" #VALUE_ATTRIBUTE="automountInformation"
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount" #MAP_ATTRIBUTE="automountMapName" #ENTRY_ATTRIBUTE="automountKey" #VALUE_ATTRIBUTE="automountInformation"
#
# AUTH_CONF_FILE - set the default location for the SASL
# authentication configuration file.
#
#AUTH_CONF_FILE="/etc/autofs_ldap_auth.conf"
#
# General global options
#
#OPTIONS=""
#

The "linux" machine is the NFS server which provides the root file system for the "lroom" and "gym" machines. The /tftpboot/muekow/i386 directory is the root file system for the diskless machines, which is a static hard mount setup during the early boot process.

Here are the log files for the "lroom" machine which works properly.

Jan  1 11:43:37 lroom automount[1968]: Starting automounter version
5.0.1-31, master map auto.master
Jan  1 11:43:37 lroom automount[1968]: using kernel protocol version 5.00
Jan  1 11:43:37 linux mountd[2130]: authenticated unmount request from
lroom:915 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 linux mountd[2130]: authenticated unmount request from
lroom:637 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 lroom automount[1968]: mounted indirect mount on
/mnt/usb with timeout 2, freq 1 seconds
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry
/mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry
/mnt/usb/dcim
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/dcim
Jan  1 11:45:17 lroom xinetd[1820]: START: telnet pid=2280

from=192.168.1.254

Here are the log files for the "gym" machine which fails to start automount properly. The automount program is in the hung state at this point.

Jan  1 11:49:36 gym automount[1846]: Starting automounter version
5.0.1-31, master map auto.master
Jan  1 11:49:36 gym automount[1846]: using kernel protocol version 5.00
Jan  1 11:49:36 linux mountd[2130]: authenticated unmount request from

gym:778 for /tftpboot/muekow/i386 (/tftpboot/muekow)

I realized today that the machines which fail to work are P3 processors whereas the one that works is a P4 processor. I don't know if that has meaning or is a pure coincidence.

Thanks for the help.

Jim

Quoted 20 lines · Ian Kent
Quoted 10 lines · Jim Duda

I'm attempting to use autofs on some diskless machines which operate using an NFS mounted root file system. I have three different computers which share the same root file system (with unique /tmp and /var directories) and the same 2.6.22 kernel.

I want to use autofs for removable USB memory sticks.

On one machine, I can use autofs just fine. Autofs mounts and unmounts the usb sticks as I expect it to. Works nice.

Quoted 8 lines

However, on the other two machines, I cannot get automount to start. It simply hangs upon startup, I cannot ^C to exit or anything. I have to kill -9 the process.

I cannot figure out what automount is looking for, and not getting, which would cause it to hang. I've attempted to use gdb to help me find the problem, but I haven't had much luck. (I've downloaded the most recent 5.0.2 tar bar and build automount with -g to debug.

Quoted 8 lines

Did you apply any of the patches for 5.0.2?

Quoted 1 line

I'm hoping someone might shed some light on how best to debug this issue.

Have a look at http://people.redhat.com/jmoyer for what to collect and post here.

Ian

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

maintainer
#5 lore raw
Quoted 3 lines · Jim Duda

My syslog.conf has this:
*.* @192.168.1.254
Doesn't that cover "everything", including daemon?

You'd think so.
All I can say is that the log is missing a bunch of info so there's something not right.

Quoted 5 lines

No core dumps.

I started using gdb to find where the program was hanging. I was down deep into reading the nisswitch.conf file, then got lost.

If you can get into gdb when autofs is wedged then how about just doing:

info threads

and

thr a a bt

Quoted 264 lines

Jim

"Ian Kent" <[email protected]> wrote in message news:[email protected]...

Quoted 6 lines · Jim Duda
Quoted 4 lines

I have gathered the information required in this post. I have provided the common information for machines which work and fail. I have provided the syslog debug message for those that work and also those that fail.

Quoted 246 lines

This all looks OK but the debug log doesn't have the debugging info. Did you set syslog to send daemon.* to the log?

I'm not sure we'll get much more from it but we should at least have a look at it.

Do you get any core files for automount?

Quoted 238 lines

lroom# rpm -q autofs
autofs-5.0.1-31

lroom# uname -a
Linux lroom 2.6.22.2-tftp #1 Sun Aug 12 11:14:47 EDT 2007 i686 i686 i386 GNU/Linux

lroom# cat /etc/auto.master
#
# Sample auto.master file
# This is an automounter map and it has the following format

# key [ -mount-options-separated-by-comma ] location
# For details of the format look at autofs(5).
#
#/misc  /etc/auto.misc
#/net   -hosts
/mnt/usb /etc/auto.usb --timeout=2
#
# Include central master map if it can be found using
# nsswitch sources.
#
# Note that if there are entries for /net or /misc (as
# above) in the included master map any keys that are the
# same will not be seen as the first read key seen takes
# precedence.
#
#+auto.master

lroom# cat /etc/auto.usb

device_a         -fstype=vfat,rw,gid=100,umask=002       :/dev/sda1
device_b         -fstype=vfat,rw,gid=100,umask=002       :/dev/sdb1

lroom# cat /etc/nsswitch.conf
#

# /etc/nsswitch.conf
#
# An example Name Service Switch config file. This file should be
# sorted with the most-used services at the beginning.
#
# The entry '[NOTFOUND=return]' means that the search for an
# entry should stop if the search in the previous entry turned
# up nothing. Note that if the search failed due to some other reason
# (like no NIS server responding) then the search continues with the
# next entry.
#
# Legal entries are:
#
#       nisplus or nis+         Use NIS+ (NIS version 3)
#       nis or yp               Use NIS (NIS version 2), also called YP
#       dns                     Use DNS (Domain Name Service)
#       files                   Use the local files
#       db                      Use the local database (.db) files
#       compat                  Use NIS on compat mode
#       hesiod                  Use Hesiod for user lookups
#       [NOTFOUND=return]       Stop searching if not found so far
#

# To use db, put the "db" in front of "files" for entries you want to be

# looked up first in the databases
#
# Example:
#passwd:    db files nisplus nis
#shadow:    db files nisplus nis
#group:     db files nisplus nis
passwd:     files
shadow:     files
group:      files

#hosts: db files nisplus nis dns
hosts: files dns

# Example - obey only what nisplus tells us...

#services:   nisplus [NOTFOUND=return] files
#networks:   nisplus [NOTFOUND=return] files
#protocols:  nisplus [NOTFOUND=return] files
#rpc:        nisplus [NOTFOUND=return] files
#ethers:     nisplus [NOTFOUND=return] files
#netmasks:   nisplus [NOTFOUND=return] files

bootparams: nisplus [NOTFOUND=return] files

ethers:     files
netmasks:   files
networks:   files
protocols:  files
rpc:        files
services:   files

netgroup: files

publickey: nisplus

automount:  files
aliases:    files nisplus

lroom# cat /etc/sysconfig/autofs
#
# Define default options for autofs.
#
# MASTER_MAP_NAME - default map name for the master map. #
#MASTER_MAP_NAME="auto.master"
#
# TIMEOUT - set the default mount timeout (default 600). #
TIMEOUT=300
#
# BROWSE_MODE - maps are browsable by default.
#
BROWSE_MODE="no"
#
# APPEND_OPTIONS - append to global options instead of replace. #
#APPEND_OPTIONS="yes"
#
# LOGGING - set default log level "none", "verbose" or "debug" #
LOGGING="debug"
#
# Define the default LDAP schema to use for lookups
#
# System default
#
#MAP_OBJECT_CLASS="nisMap" #ENTRY_OBJECT_CLASS="nisObject" #MAP_ATTRIBUTE="nisMapName"
#ENTRY_ATTRIBUTE="cn"
#VALUE_ATTRIBUTE="nisMapEntry"
#
# Other common LDAP nameing
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount"
#MAP_ATTRIBUTE="ou"
#ENTRY_ATTRIBUTE="cn" #VALUE_ATTRIBUTE="automountInformation"
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount" #MAP_ATTRIBUTE="automountMapName" #ENTRY_ATTRIBUTE="automountKey" #VALUE_ATTRIBUTE="automountInformation"
#
# AUTH_CONF_FILE - set the default location for the SASL
# authentication configuration file.
#
#AUTH_CONF_FILE="/etc/autofs_ldap_auth.conf"
#
# General global options
#
#OPTIONS=""
#

The "linux" machine is the NFS server which provides the root file system for the "lroom" and "gym" machines. The /tftpboot/muekow/i386 directory is the root file system for the diskless machines, which is a static hard mount setup during the early boot process.

Here are the log files for the "lroom" machine which works properly.

Jan  1 11:43:37 lroom automount[1968]: Starting automounter version
5.0.1-31, master map auto.master
Jan  1 11:43:37 lroom automount[1968]: using kernel protocol version 5.00
Jan  1 11:43:37 linux mountd[2130]: authenticated unmount request from
lroom:915 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 linux mountd[2130]: authenticated unmount request from
lroom:637 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 lroom automount[1968]: mounted indirect mount on
/mnt/usb with timeout 2, freq 1 seconds
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry
/mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry
/mnt/usb/dcim
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/dcim
Jan  1 11:45:17 lroom xinetd[1820]: START: telnet pid=2280

from=192.168.1.254

Here are the log files for the "gym" machine which fails to start automount properly. The automount program is in the hung state at this point.

Jan  1 11:49:36 gym automount[1846]: Starting automounter version
5.0.1-31, master map auto.master
Jan  1 11:49:36 gym automount[1846]: using kernel protocol version 5.00
Jan  1 11:49:36 linux mountd[2130]: authenticated unmount request from

gym:778 for /tftpboot/muekow/i386 (/tftpboot/muekow)

I realized today that the machines which fail to work are P3 processors whereas the one that works is a P4 processor. I don't know if that has meaning or is a pure coincidence.

Thanks for the help.

Jim

Quoted 20 lines · Ian Kent
Quoted 10 lines · Jim Duda

I'm attempting to use autofs on some diskless machines which operate using an NFS mounted root file system. I have three different computers which share the same root file system (with unique /tmp and /var directories) and the same 2.6.22 kernel.

I want to use autofs for removable USB memory sticks.

On one machine, I can use autofs just fine. Autofs mounts and unmounts the usb sticks as I expect it to. Works nice.

Quoted 8 lines

However, on the other two machines, I cannot get automount to start. It simply hangs upon startup, I cannot ^C to exit or anything. I have to kill -9 the process.

I cannot figure out what automount is looking for, and not getting, which would cause it to hang. I've attempted to use gdb to help me find the problem, but I haven't had much luck. (I've downloaded the most recent 5.0.2 tar bar and build automount with -g to debug.

Quoted 8 lines

Did you apply any of the patches for 5.0.2?

Quoted 1 line

I'm hoping someone might shed some light on how best to debug this issue.

Have a look at http://people.redhat.com/jmoyer for what to collect and post here.

Ian

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

author
#6 lore raw

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I'm by no means a power gdb user.

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

Again, one machine works fine while two others get wedged.

Jim

Quoted 279 lines · Ian Kent
Quoted 3 lines · Jim Duda

My syslog.conf has this:
*.* @192.168.1.254
Doesn't that cover "everything", including daemon?

You'd think so.
All I can say is that the log is missing a bunch of info so there's something not right.

Quoted 4 lines

No core dumps.

I started using gdb to find where the program was hanging. I was down deep into reading the nisswitch.conf file, then got lost.

If you can get into gdb when autofs is wedged then how about just doing:

info threads

and

thr a a bt

Quoted 257 lines

Jim

"Ian Kent" <[email protected]> wrote in message news:[email protected]...

Quoted 5 lines · Jim Duda
Quoted 4 lines

I have gathered the information required in this post. I have provided the common information for machines which work and fail. I have provided the syslog debug message for those that work and also those that fail.

Quoted 242 lines

This all looks OK but the debug log doesn't have the debugging info. Did you set syslog to send daemon.* to the log?

I'm not sure we'll get much more from it but we should at least have a look at it.

Do you get any core files for automount?

Quoted 234 lines

lroom# rpm -q autofs
autofs-5.0.1-31

lroom# uname -a
Linux lroom 2.6.22.2-tftp #1 Sun Aug 12 11:14:47 EDT 2007 i686 i686 i386 GNU/Linux

lroom# cat /etc/auto.master
#
# Sample auto.master file
# This is an automounter map and it has the following format

# key [ -mount-options-separated-by-comma ] location
# For details of the format look at autofs(5).
#
#/misc  /etc/auto.misc
#/net   -hosts
/mnt/usb /etc/auto.usb --timeout=2
#
# Include central master map if it can be found using
# nsswitch sources.
#
# Note that if there are entries for /net or /misc (as
# above) in the included master map any keys that are the
# same will not be seen as the first read key seen takes
# precedence.
#
#+auto.master

lroom# cat /etc/auto.usb

device_a         -fstype=vfat,rw,gid=100,umask=002       :/dev/sda1
device_b         -fstype=vfat,rw,gid=100,umask=002       :/dev/sdb1

lroom# cat /etc/nsswitch.conf
#

# /etc/nsswitch.conf
#
# An example Name Service Switch config file. This file should be
# sorted with the most-used services at the beginning.
#
# The entry '[NOTFOUND=return]' means that the search for an
# entry should stop if the search in the previous entry turned
# up nothing. Note that if the search failed due to some other reason
# (like no NIS server responding) then the search continues with the
# next entry.
#
# Legal entries are:
#
#       nisplus or nis+         Use NIS+ (NIS version 3)
#       nis or yp               Use NIS (NIS version 2), also called YP
#       dns                     Use DNS (Domain Name Service)
#       files                   Use the local files
#       db                      Use the local database (.db) files
#       compat                  Use NIS on compat mode
#       hesiod                  Use Hesiod for user lookups
#       [NOTFOUND=return]       Stop searching if not found so far
#

# To use db, put the "db" in front of "files" for entries you want to be

# looked up first in the databases
#
# Example:
#passwd:    db files nisplus nis
#shadow:    db files nisplus nis
#group:     db files nisplus nis
passwd:     files
shadow:     files
group:      files

#hosts: db files nisplus nis dns
hosts: files dns

# Example - obey only what nisplus tells us...

#services:   nisplus [NOTFOUND=return] files
#networks:   nisplus [NOTFOUND=return] files
#protocols:  nisplus [NOTFOUND=return] files
#rpc:        nisplus [NOTFOUND=return] files
#ethers:     nisplus [NOTFOUND=return] files
#netmasks:   nisplus [NOTFOUND=return] files

bootparams: nisplus [NOTFOUND=return] files

ethers:     files
netmasks:   files
networks:   files
protocols:  files
rpc:        files
services:   files

netgroup: files

publickey: nisplus

automount:  files
aliases:    files nisplus

lroom# cat /etc/sysconfig/autofs
#
# Define default options for autofs.
#
# MASTER_MAP_NAME - default map name for the master map. #
#MASTER_MAP_NAME="auto.master"
#
# TIMEOUT - set the default mount timeout (default 600). #
TIMEOUT=300
#
# BROWSE_MODE - maps are browsable by default.
#
BROWSE_MODE="no"
#
# APPEND_OPTIONS - append to global options instead of replace. #
#APPEND_OPTIONS="yes"
#
# LOGGING - set default log level "none", "verbose" or "debug" #
LOGGING="debug"
#
# Define the default LDAP schema to use for lookups
#
# System default
#
#MAP_OBJECT_CLASS="nisMap" #ENTRY_OBJECT_CLASS="nisObject" #MAP_ATTRIBUTE="nisMapName"
#ENTRY_ATTRIBUTE="cn"
#VALUE_ATTRIBUTE="nisMapEntry"
#
# Other common LDAP nameing
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount"
#MAP_ATTRIBUTE="ou"
#ENTRY_ATTRIBUTE="cn" #VALUE_ATTRIBUTE="automountInformation"
#
#MAP_OBJECT_CLASS="automountMap" #ENTRY_OBJECT_CLASS="automount" #MAP_ATTRIBUTE="automountMapName" #ENTRY_ATTRIBUTE="automountKey" #VALUE_ATTRIBUTE="automountInformation"
#
# AUTH_CONF_FILE - set the default location for the SASL
# authentication configuration file.
#
#AUTH_CONF_FILE="/etc/autofs_ldap_auth.conf"
#
# General global options
#
#OPTIONS=""
#

The "linux" machine is the NFS server which provides the root file system for the "lroom" and "gym" machines. The /tftpboot/muekow/i386 directory is the root file system for the diskless machines, which is a static hard mount setup during the early boot process.

Here are the log files for the "lroom" machine which works properly.

Jan  1 11:43:37 lroom automount[1968]: Starting automounter version
5.0.1-31, master map auto.master
Jan  1 11:43:37 lroom automount[1968]: using kernel protocol version 5.00
Jan  1 11:43:37 linux mountd[2130]: authenticated unmount request from
lroom:915 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 linux mountd[2130]: authenticated unmount request from
lroom:637 for /tftpboot/muekow/i386 (/tftpboot/muekow)
Jan  1 11:44:12 lroom automount[1968]: mounted indirect mount on
/mnt/usb with timeout 2, freq 1 seconds
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry
/mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/DCIM
Jan  1 11:45:06 lroom automount[1968]: attempting to mount entry
/mnt/usb/dcim
Jan  1 11:45:06 lroom automount[1968]: failed to mount /mnt/usb/dcim
Jan  1 11:45:17 lroom xinetd[1820]: START: telnet pid=2280

from=192.168.1.254

Here are the log files for the "gym" machine which fails to start automount properly. The automount program is in the hung state at this point.

Jan  1 11:49:36 gym automount[1846]: Starting automounter version
5.0.1-31, master map auto.master
Jan  1 11:49:36 gym automount[1846]: using kernel protocol version 5.00
Jan  1 11:49:36 linux mountd[2130]: authenticated unmount request from

gym:778 for /tftpboot/muekow/i386 (/tftpboot/muekow)

I realized today that the machines which fail to work are P3 processors whereas the one that works is a P4 processor. I don't know if that has meaning or is a pure coincidence.

Thanks for the help.

Jim

Quoted 19 lines · Ian Kent
Quoted 10 lines · Jim Duda

I'm attempting to use autofs on some diskless machines which operate using an NFS mounted root file system. I have three different computers which share the same root file system (with unique /tmp and /var directories) and the same 2.6.22 kernel.

I want to use autofs for removable USB memory sticks.

On one machine, I can use autofs just fine. Autofs mounts and unmounts the usb sticks as I expect it to. Works nice.

Quoted 8 lines

However, on the other two machines, I cannot get automount to start. It simply hangs upon startup, I cannot ^C to exit or anything. I have to kill -9 the process.

I cannot figure out what automount is looking for, and not getting, which would cause it to hang. I've attempted to use gdb to help me find the problem, but I haven't had much luck. (I've downloaded the most recent 5.0.2 tar bar and build automount with -g to debug.

Quoted 7 lines

Did you apply any of the patches for 5.0.2?

Quoted 1 line

I'm hoping someone might shed some light on how best to debug this issue.

Have a look at http://people.redhat.com/jmoyer for what to collect and post here.

Ian

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

maintainer
#7 lore raw
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 5 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I'm by no means a power gdb user.

Me nether.

Quoted 10 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221

autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
diff -u -r1.221 autofs.spec
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
 %build
author
#8 lore raw

Ian,

I do have daemon.*, I got it backwards in the last post.

I downloaded autofs-5.0.2.tar.gz, do you want me to download 5.1.31 ?

Jim

Quoted 65 lines · Ian Kent
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 4 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I'm by no means a power gdb user.

Me nether.

Quoted 9 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221
diff -u -r1.221 autofs.spec

autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
 %build
author
#9 lore raw

Ian,

Here in spawn.c

		errp = 0;
		do {
			while ((errn =
				read(pipefd[0], errbuf + errp, ERRBUFSIZ - errp)) == -1
			       && errno == EINTR);
Quoted 71 lines · Jim Duda

Ian,

I do have daemon.*, I got it backwards in the last post.

I downloaded autofs-5.0.2.tar.gz, do you want me to download 5.1.31 ?

Jim

Quoted 62 lines · Ian Kent
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 4 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I'm by no means a power gdb user.

Me nether.

Quoted 9 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221
diff -u -r1.221 autofs.spec

autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
 %build
maintainer
#10 lore raw
Quoted 9 lines · Jim Duda

Ian,

Here in spawn.c

		errp = 0;
		do {
			while ((errn =
				read(pipefd[0], errbuf + errp, ERRBUFSIZ - errp)) == -1
			       && errno == EINTR);

I'm not sure this really says much except that autofs is waiting for mount(8) (or umount(8) as the case may be) to complete.

Usually you will see a mount process running if mount isn't completing, in which case, it becomes a matter of working out why mount can't complete the mount. If we were logging debug info then usually we would be able to get the mount command used which is often helpful. OTOH, if there isn't any process (either mount or another automount process) then perhaps mount has seg faulted and autofs is waiting for a reply but, usually, if that happens the daemon gets a signal and continues with what looks like a successful mount which it often really isn't.

Quoted 21 lines
Quoted 18 lines · Jim Duda

Ian,

I do have daemon.*, I got it backwards in the last post.

I downloaded autofs-5.0.2.tar.gz, do you want me to download 5.1.31 ?

Jim

Quoted 9 lines · Ian Kent
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 2 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I haven't seen that behavior before and I've debugged some really broken code in the early version 5 development. Perhaps this is something other than just autofs?

Quoted 53 lines
Quoted 47 lines
Quoted 47 lines
Quoted 2 lines

I'm by no means a power gdb user.

Me nether.

Quoted 9 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221
diff -u -r1.221 autofs.spec

autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
Quoted 1 line

autofs.spec 4 Jan 2008 02:20:20 -0000 line 133

Quoted 1 line
autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +0 -0
 %build

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

author
#11 lore raw

Ian,

On the machine that don't work properly, I get this from showmount.

asterisk> showmount
Broken pipe

Does that mean anything to you?

Note that in my auto.master file, I have /net commented out ...

Jim

Quoted 104 lines · Ian Kent
Quoted 9 lines · Jim Duda

Ian,

Here in spawn.c

		errp = 0;
		do {
			while ((errn =
				read(pipefd[0], errbuf + errp, ERRBUFSIZ - errp)) == -1
			       && errno == EINTR);

I'm not sure this really says much except that autofs is waiting for mount(8) (or umount(8) as the case may be) to complete.

Usually you will see a mount process running if mount isn't completing, in which case, it becomes a matter of working out why mount can't complete the mount. If we were logging debug info then usually we would be able to get the mount command used which is often helpful. OTOH, if there isn't any process (either mount or another automount process) then perhaps mount has seg faulted and autofs is waiting for a reply but, usually, if that happens the daemon gets a signal and continues with what looks like a successful mount which it often really isn't.

Quoted 20 lines
Quoted 18 lines · Jim Duda

Ian,

I do have daemon.*, I got it backwards in the last post.

I downloaded autofs-5.0.2.tar.gz, do you want me to download 5.1.31 ?

Jim

Quoted 9 lines · Ian Kent
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 2 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I haven't seen that behavior before and I've debugged some really broken code in the early version 5 development. Perhaps this is something other than just autofs?

Quoted 51 lines
Quoted 46 lines
Quoted 46 lines
Quoted 1 line

I'm by no means a power gdb user.

Me nether.

Quoted 9 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221
diff -u -r1.221 autofs.spec

autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
Quoted 1 line

autofs.spec 4 Jan 2008 02:20:20 -0000 line 133

Quoted 1 line
autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +0 -0
 %build

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

maintainer
#12 lore raw
Quoted 6 lines · Jim Duda

Ian,

On the machine that don't work properly, I get this from showmount.

asterisk> showmount
Broken pipe

That doesn't look very good.
It may be an indication that RPC is somehow not right on the machine. I would expect output like (assuming the machine isn't actually being used as an NFS server):
showmount: RPC: Program not registered

Maybe the problem is as simple as portmap or rpcbind not running on the machine. But then I'd expect showmount to just return nothing.

Quoted 113 lines

Does that mean anything to you?

Note that in my auto.master file, I have /net commented out ...

Jim

Quoted 104 lines · Ian Kent
Quoted 9 lines · Jim Duda

Ian,

Here in spawn.c

		errp = 0;
		do {
			while ((errn =
				read(pipefd[0], errbuf + errp, ERRBUFSIZ - errp)) == -1
			       && errno == EINTR);

I'm not sure this really says much except that autofs is waiting for mount(8) (or umount(8) as the case may be) to complete.

Usually you will see a mount process running if mount isn't completing, in which case, it becomes a matter of working out why mount can't complete the mount. If we were logging debug info then usually we would be able to get the mount command used which is often helpful. OTOH, if there isn't any process (either mount or another automount process) then perhaps mount has seg faulted and autofs is waiting for a reply but, usually, if that happens the daemon gets a signal and continues with what looks like a successful mount which it often really isn't.

Quoted 20 lines
Quoted 18 lines · Jim Duda

Ian,

I do have daemon.*, I got it backwards in the last post.

I downloaded autofs-5.0.2.tar.gz, do you want me to download 5.1.31 ?

Jim

Quoted 9 lines · Ian Kent
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 2 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I haven't seen that behavior before and I've debugged some really broken code in the early version 5 development. Perhaps this is something other than just autofs?

Quoted 51 lines
Quoted 46 lines
Quoted 46 lines
Quoted 1 line

I'm by no means a power gdb user.

Me nether.

Quoted 9 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221
diff -u -r1.221 autofs.spec

autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
Quoted 1 line

autofs.spec 4 Jan 2008 02:20:20 -0000 line 133

Quoted 1 line
autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +0 -0
 %build

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

author
#13 lore raw

Ian,

I cannot explain why, but I wasn't able to get any debug message via SYSLOG, however, I found that by using msg(""), I was able to add my own debugging.

I finally figured out the root cause of the hang. The program was blocked down inside replicate.d, at the read of /dev/random.

I have both /dev/random and /dev/urandom.

By simply replacing /dev/random with /dev/urandom in replicate.c, my hang problem was resolved and automount works just great.

Does that make any sense?

Jim

Quoted 125 lines · Ian Kent
Quoted 6 lines · Jim Duda

Ian,

On the machine that don't work properly, I get this from showmount.

asterisk> showmount
Broken pipe

That doesn't look very good.
It may be an indication that RPC is somehow not right on the machine. I would expect output like (assuming the machine isn't actually being used as an NFS server):
showmount: RPC: Program not registered

Maybe the problem is as simple as portmap or rpcbind not running on the machine. But then I'd expect showmount to just return nothing.

Quoted 108 lines

Does that mean anything to you?

Note that in my auto.master file, I have /net commented out ...

Jim

Quoted 101 lines · Ian Kent
Quoted 9 lines · Jim Duda

Ian,

Here in spawn.c

		errp = 0;
		do {
			while ((errn =
				read(pipefd[0], errbuf + errp, ERRBUFSIZ - errp)) == -1
			       && errno == EINTR);

I'm not sure this really says much except that autofs is waiting for mount(8) (or umount(8) as the case may be) to complete.

Usually you will see a mount process running if mount isn't completing, in which case, it becomes a matter of working out why mount can't complete the mount. If we were logging debug info then usually we would be able to get the mount command used which is often helpful. OTOH, if there isn't any process (either mount or another automount process) then perhaps mount has seg faulted and autofs is waiting for a reply but, usually, if that happens the daemon gets a signal and continues with what looks like a successful mount which it often really isn't.

Quoted 19 lines
Quoted 18 lines · Jim Duda

Ian,

I do have daemon.*, I got it backwards in the last post.

I downloaded autofs-5.0.2.tar.gz, do you want me to download 5.1.31 ?

Jim

Quoted 9 lines · Ian Kent
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 2 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I haven't seen that behavior before and I've debugged some really broken code in the early version 5 development. Perhaps this is something other than just autofs?

Quoted 51 lines
Quoted 46 lines
Quoted 46 lines
Quoted 1 line

I'm by no means a power gdb user.

Me nether.

Quoted 9 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221
diff -u -r1.221 autofs.spec

autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
Quoted 1 line

autofs.spec 4 Jan 2008 02:20:20 -0000 line 133

Quoted 1 line
autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +0 -0
 %build

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

author
#14 lore raw

Ian,

All of my NFS based root file system computers operate without mice or keyboards. I read a few posts which claim that /dev/random might not provide any data on these type of machines.

In my rc.local, I have simply removed /dev/random and created a link from random to /dev/urandom.

Now everything works just fine!!

Thanks for all your patience on this thread.

Jim

Quoted 143 lines · Jim Duda

Ian,

I cannot explain why, but I wasn't able to get any debug message via SYSLOG, however, I found that by using msg(""), I was able to add my own debugging.

I finally figured out the root cause of the hang. The program was blocked down inside replicate.d, at the read of /dev/random.

I have both /dev/random and /dev/urandom.

By simply replacing /dev/random with /dev/urandom in replicate.c, my hang problem was resolved and automount works just great.

Does that make any sense?

Jim

Quoted 124 lines · Ian Kent
Quoted 6 lines · Jim Duda

Ian,

On the machine that don't work properly, I get this from showmount.

asterisk> showmount
Broken pipe

That doesn't look very good.
It may be an indication that RPC is somehow not right on the machine. I would expect output like (assuming the machine isn't actually being used as an NFS server):
showmount: RPC: Program not registered

Maybe the problem is as simple as portmap or rpcbind not running on the machine. But then I'd expect showmount to just return nothing.

Quoted 108 lines

Does that mean anything to you?

Note that in my auto.master file, I have /net commented out ...

Jim

Quoted 101 lines · Ian Kent
Quoted 9 lines · Jim Duda

Ian,

Here in spawn.c

		errp = 0;
		do {
			while ((errn =
				read(pipefd[0], errbuf + errp, ERRBUFSIZ - errp)) == -1
			       && errno == EINTR);

I'm not sure this really says much except that autofs is waiting for mount(8) (or umount(8) as the case may be) to complete.

Usually you will see a mount process running if mount isn't completing, in which case, it becomes a matter of working out why mount can't complete the mount. If we were logging debug info then usually we would be able to get the mount command used which is often helpful. OTOH, if there isn't any process (either mount or another automount process) then perhaps mount has seg faulted and autofs is waiting for a reply but, usually, if that happens the daemon gets a signal and continues with what looks like a successful mount which it often really isn't.

Quoted 19 lines
Quoted 18 lines · Jim Duda

Ian,

I do have daemon.*, I got it backwards in the last post.

I downloaded autofs-5.0.2.tar.gz, do you want me to download 5.1.31 ?

Jim

Quoted 9 lines · Ian Kent
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 2 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I haven't seen that behavior before and I've debugged some really broken code in the early version 5 development. Perhaps this is something other than just autofs?

Quoted 51 lines
Quoted 46 lines
Quoted 46 lines
Quoted 1 line

I'm by no means a power gdb user.

Me nether.

Quoted 9 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221
diff -u -r1.221 autofs.spec

autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
Quoted 1 line

autofs.spec 4 Jan 2008 02:20:20 -0000 line 133

Quoted 1 line
autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +0 -0
 %build

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

maintainer
#15 lore raw
Quoted 6 lines · Jim Duda

Ian,

All of my NFS based root file system computers operate without mice or keyboards. I read a few posts which claim that /dev/random might not provide any data on these type of machines.

Quoted 6 lines

In my rc.local, I have simply removed /dev/random and created a link from random to /dev/urandom.

Now everything works just fine!!

Thanks for all your patience on this thread.

Wow, that's a surprise and yet so obvious in hindsight. I'll have to work out a way to deal with that.

Quoted 148 lines

Jim

Quoted 143 lines · Jim Duda

Ian,

I cannot explain why, but I wasn't able to get any debug message via SYSLOG, however, I found that by using msg(""), I was able to add my own debugging.

I finally figured out the root cause of the hang. The program was blocked down inside replicate.d, at the read of /dev/random.

I have both /dev/random and /dev/urandom.

By simply replacing /dev/random with /dev/urandom in replicate.c, my hang problem was resolved and automount works just great.

Does that make any sense?

Jim

Quoted 124 lines · Ian Kent
Quoted 6 lines · Jim Duda

Ian,

On the machine that don't work properly, I get this from showmount.

asterisk> showmount
Broken pipe

That doesn't look very good.
It may be an indication that RPC is somehow not right on the machine. I would expect output like (assuming the machine isn't actually being used as an NFS server):
showmount: RPC: Program not registered

Maybe the problem is as simple as portmap or rpcbind not running on the machine. But then I'd expect showmount to just return nothing.

Quoted 108 lines

Does that mean anything to you?

Note that in my auto.master file, I have /net commented out ...

Jim

Quoted 101 lines · Ian Kent
Quoted 9 lines · Jim Duda

Ian,

Here in spawn.c

		errp = 0;
		do {
			while ((errn =
				read(pipefd[0], errbuf + errp, ERRBUFSIZ - errp)) == -1
			       && errno == EINTR);

I'm not sure this really says much except that autofs is waiting for mount(8) (or umount(8) as the case may be) to complete.

Usually you will see a mount process running if mount isn't completing, in which case, it becomes a matter of working out why mount can't complete the mount. If we were logging debug info then usually we would be able to get the mount command used which is often helpful. OTOH, if there isn't any process (either mount or another automount process) then perhaps mount has seg faulted and autofs is waiting for a reply but, usually, if that happens the daemon gets a signal and continues with what looks like a successful mount which it often really isn't.

Quoted 19 lines
Quoted 18 lines · Jim Duda

Ian,

I do have daemon.*, I got it backwards in the last post.

I downloaded autofs-5.0.2.tar.gz, do you want me to download 5.1.31 ?

Jim

Quoted 9 lines · Ian Kent
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 2 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I haven't seen that behavior before and I've debugged some really broken code in the early version 5 development. Perhaps this is something other than just autofs?

Quoted 51 lines
Quoted 46 lines
Quoted 46 lines
Quoted 1 line

I'm by no means a power gdb user.

Me nether.

Quoted 9 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221
diff -u -r1.221 autofs.spec

autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
Quoted 1 line

autofs.spec 4 Jan 2008 02:20:20 -0000 line 133

Quoted 1 line
autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +0 -0
 %build

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

maintainer
#16 lore raw
In reply to Jim Duda
Quoted 5 lines · Jim Duda

Ian,

I do have daemon.*, I got it backwards in the last post.

I downloaded autofs-5.0.2.tar.gz, do you want me to download 5.1.31 ?

I only said that because you mentioned using autofs-5.0.1-31 in a previous mail. You did mention you had downloaded the tar but didn't say if you had applied the patches when I asked and in a mail following you provided output from "rpm -q autofs" that showed you were using revision 31.

The patch I referred to isn't included in the tar or any of the patches that should be applied to it.

Quoted 74 lines

Jim

Quoted 57 lines · Ian Kent
Quoted 4 lines · Jim Duda

Ian,

Adding *.daemon simply resulted in the same information being dumped to the syslog file, however, twice. So, no new information.

That should be daemon.* and usually you would log it to a different file when adding a syslog entry like that but I don't think that will make any difference.

I can't remember how logging to a syslog server works now but does the syslog configuration on the server also limit what is logged?

Quoted 4 lines

Once automount gets wedged, I cannot use gdb to interrogate the threads, I cannot break into the program after it's wedged.

I'm by no means a power gdb user.

Me nether.

Quoted 9 lines

I did:

set detach-on-fork off, simply based on a recommended help from ddd.

I traced the program all the way down into mount_bind.c, in the mount_init function, then into spawn.c, where it did the first fork. The program was wedged in spawn.c on line 186 at the first do while loop after the fork.

Quoted 6 lines

The program though do_read_master, mod->lookup_int, then into open_mount for "nfs" before it got to the first spawn.

I don't know how helpful any of this information is for you in helping me determine what is different about my funky root file system which causes a lockup, but thanks for trying.

I'm not sure either but one thing is sure, problems are almost always different from what you think they are when you finally get hard evidence.

In autofs-5.0.1-31, line 186 corresponds to an if statement?

How about we try getting rid of a recent patch to this area of the code and rebuild autofs and see if that helps. The one I have in mind is close to the chopping block already.

In particular:

[raven@raven F-7]$ cvs diff -u autofs.spec
Index: autofs.spec
===================================================================
RCS file: /cvs/pkgs/rpms/autofs/F-7/autofs.spec,v
retrieving revision 1.221
diff -u -r1.221 autofs.spec

autofs.spec 4 Jan 2008 02:20:20 -0000 line 133

Quoted 8 lines
autofs.spec 4 Jan 2008 02:20:20 -0000 renamed from autofs.spec 21 Dec 2007 10:21:18 -0000 1.221 +1 -1
@@ -127,7 +127,7 @@
 %patch35 -p1
 %patch36 -p1
 %patch37 -p1
-%patch38 -p1
+#%patch38 -p1
 %patch39 -p1
 
 %build

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs