Skip to contentSkip to conversation

2.4.x end of tape handling error

DISCUSSION Kevin Fenzi 2 participants 3 messages Feb 10, 2003 – Feb 18, 2003 SCSI

lore
Share & tools
lore
#0 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Greetings.

I have had reported from a client that they are having problems with backups that span more than one tape. Instead of getting an EOT error or EOM, they are getting an I/O error wich requires the driver to be unloaded and reloaded before the tape will work again.

http://www.linuxtapecert.org/ Says that the redhat 2.4.9-34 kernel is the last one that had proper EOT handling. Indeed, if they use the 2.4.9-34 kernel, the tape works properly. Thats not a very good solution however.

Is this fixed in the latest 2.4.21-pres? How about in 2.5.x?

Has anyone else seen this?

I can get further details to track this down and fix it if it's not already fixed.

kevin
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>

iD8DBQE+R+3+3imCezTjY0ERAuZMAKCcQJInxbLBaOOFaTdIRxFhzVT+LQCfSqHg 9HpLrQOVYetev4zhYXnFD/o=
=WN5h
-----END PGP SIGNATURE-----

#1 lore raw

This discussion should really be moved to linux-scsi...

Quoted 7 lines · Kevin Fenzi

Greetings.

I have had reported from a client that they are having problems with backups that span more than one tape. Instead of getting an EOT error or EOM, they are getting an I/O error wich requires the driver to be unloaded and reloaded before the tape will work again.

What messages have they seen in the system log? Some messages should be after this kind of error. It is difficult to see where the problem is without any details. There have not been any significant changes in EOM handling in st between the 2.4 kernels.

Quoted 7 lines

http://www.linuxtapecert.org/ Says that the redhat 2.4.9-34 kernel is the last one that had proper EOT handling. Indeed, if they use the 2.4.9-34 kernel, the tape works properly. Thats not a very good solution however.

Is this fixed in the latest 2.4.21-pres? How about in 2.5.x?

Don't know. EOM handling has worked with my test system (HP DDS drives connected to a SYM53c896) with both 2.4 and 2.5 kernels. I just reran the eom tests with 2.4.20 and 2.5.60 without problems.

Kai

#2 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Kai" == Kai Makisara <[email protected]> writes:

Kai> This discussion should really be moved to linux-scsi...

ok.

Quoted 7 lines · Kevin Fenzi
Quoted 7 lines

Greetings.

I have had reported from a client that they are having problems with backups that span more than one tape. Instead of getting an EOT error or EOM, they are getting an I/O error wich requires the driver to be unloaded and reloaded before the tape will work again.

Kai> What messages have they seen in the system log? Some messages Kai> should be after this kind of error. It is difficult to see where Kai> the problem is without any details. There have not been any Kai> significant changes in EOM handling in st between the 2.4 Kai> kernels.

I have duplicated the problem with a dds2 drive I have here.

Basically, they get back a EIO error instead of a ENOSPC error when hitting the end of tape.

Quoted 7 lines
Quoted 7 lines

http://www.linuxtapecert.org/ Says that the redhat 2.4.9-34 kernel is the last one that had proper EOT handling. Indeed, if they use the 2.4.9-34 kernel, the tape works properly. Thats not a very good solution however.

Is this fixed in the latest 2.4.21-pres? How about in 2.5.x?

Kai> Don't know. EOM handling has worked with my test system (HP DDS Kai> drives connected to a SYM53c896) with both 2.4 and 2.5 kernels. I Kai> just reran the eom tests with 2.4.20 and 2.5.60 without problems.

My test system is very similar.

HP DDS2 drive, on a SYM53c896. ;)
Redhat 7.3 machine, running 2.4.18-24.7.x.

To test I just used tar and tried to put more on the tape than would fit.

I got:

...
write(3, "\305\230\240=:-v\t\237&\16\317\234+x\"N(\375\0\264\000"..., 10240) = -1 EIO (Input/output error)
write(2, "tar: ", 5) = 5
write(2, "/dev/nst0: Wrote only 0 of 10240"..., 38) = 38
write(2, "\n", 1) = 1
write(2, "tar: ", 5) = 5
write(2, "Error is not recoverable: exitin"..., 37) = 37

write(2, "\n", 1)                       = 1
munmap(0x4002e000, 4096)                = 0
_exit(2)                                = ?

I tried loading the driver with various options, in case it was a buffering issue:

modprobe st buffer_kbs=0 write_threshold_kbs=0 max_buffers=0 max_sg_segs=0

Any option I seemed to give it resulted in the same thing. I would be happy to test anything or provide you access to the box with the tape drive on it.
Will also try the 2.4.9-34 kernel that is supposed to not have this issue.

Kai> Kai

kevin

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>

iD8DBQE+UaWe3imCezTjY0ERAt84AJ9JIzGEwm5H9Id4Tg4k0/o0fzZmgwCgk3zr bP+qoI82jXtyMY7XU2YHjWU=
=4Z2L
-----END PGP SIGNATURE-----