Skip to contentSkip to conversation

nfsstat --sleep=#

DISCUSSION Kevin Constantine 5 participants 18 messages Mar 12, 2009 – Mar 16, 2009 3 messages not in the archive

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference between 3 seconds ago and now. Frequently I'll run a test and want to watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

       total:        0     3476
        null:        0        0
     getattr:        0     1679
     setattr:        0        0
      lookup:        0      839
      access:        0      839
    readlink:        0        0
        read:        0        0
       write:        0        0
      create:        0        0
       mkdir:        0        0
     symlink:        0        0
       mknod:        0        0
      remove:        0        0
       rmdir:        0        0
      rename:        0        0
        link:        0        0
     readdir:        0        0
 readdirplus:        0        0
      fsstat:        0      119
      fsinfo:        0        0
    pathconf:        0        0
      commit:        0        0
#2 lore raw
In reply to a message not in the archive
Quoted 7 lines · Kevin Constantine

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference between 3 seconds ago and now. Frequently I'll run a test and want to watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I wrote some patches a couple of years back to do that. I'll see if I can dig 'em up.

Signature · 3 lines
Greg Banks, P.Engineer, SGI Australian Software Group.
the brightly coloured sporks of revolution.
I don't speak for SGI.
#3 lore raw

It seems like everyone has a patch or a script that does this sort of thing. It'd be great if it was just a part of nfsstat the way that vmstat, and iostat print differences on a regular basis.

Quoted 12 lines · Greg Banks
Quoted 7 lines · Kevin Constantine

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference between 3 seconds ago and now. Frequently I'll run a test and want to watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I wrote some patches a couple of years back to do that. I'll see if I can dig 'em up.

Signature · 8 lines
------------------------------------------------------------
Kevin Constantine

Systems Engineer		t: 818.460.8221
Walt Disney Animation Studios	e: kevin.constantine-P5ys19MLBK/[email protected]

Any sufficiently advanced technology is indistinguishable from magic.
    - Arthur C. Clarke
#4 lore raw
In reply to a message not in the archive

Hi Kevin-

man watch(1)

Quoted 37 lines · Kevin Constantine

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference between 3 seconds ago and now. Frequently I'll run a test and want to watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

      total:        0     3476
       null:        0        0
    getattr:        0     1679
    setattr:        0        0
     lookup:        0      839
     access:        0      839
   readlink:        0        0
       read:        0        0
      write:        0        0
     create:        0        0
      mkdir:        0        0
    symlink:        0        0
      mknod:        0        0
     remove:        0        0
      rmdir:        0        0
     rename:        0        0
       link:        0        0
    readdir:        0        0
readdirplus:        0        0
     fsstat:        0      119
     fsinfo:        0        0
   pathconf:        0        0
     commit:        0        0

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 2 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com
#5 lore raw
Quoted 3 lines · Chuck Lever

Hi Kevin-

man watch(1)

What would you watch?

--b.

Quoted 56 lines · Kevin Constantine
Quoted 42 lines

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference between 3 seconds ago and now. Frequently I'll run a test and want to watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

      total:        0     3476
       null:        0        0
    getattr:        0     1679
    setattr:        0        0
     lookup:        0      839
     access:        0      839
   readlink:        0        0
       read:        0        0
      write:        0        0
     create:        0        0
      mkdir:        0        0
    symlink:        0        0
      mknod:        0        0
     remove:        0        0
      rmdir:        0        0
     rename:        0        0
       link:        0        0
    readdir:        0        0
readdirplus:        0        0
     fsstat:        0      119
     fsinfo:        0        0
   pathconf:        0        0
     commit:        0        0

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
#6 lore raw
Quoted 6 lines · J. Bruce Fields
Quoted 3 lines · Chuck Lever

Hi Kevin-

man watch(1)

What would you watch?

For example:

watch -n3 nfsstat -c

You can also use the "-d" option to highlight the differences between the current sample and the previous sample.

Quoted 58 lines

--b.

Quoted 56 lines · Kevin Constantine
Quoted 42 lines

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference between 3 seconds ago and now. Frequently I'll run a test and want to watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

     total:        0     3476
      null:        0        0
   getattr:        0     1679
   setattr:        0        0
    lookup:        0      839
    access:        0      839
  readlink:        0        0
      read:        0        0
     write:        0        0
    create:        0        0
     mkdir:        0        0
   symlink:        0        0
     mknod:        0        0
    remove:        0        0
     rmdir:        0        0
    rename:        0        0
      link:        0        0
   readdir:        0        0
readdirplus:        0        0
    fsstat:        0      119
    fsinfo:        0        0
  pathconf:        0        0
    commit:        0        0

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 2 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com
#20 lore raw
In reply to Chuck Lever
Quoted 3 lines · Chuck Lever

Hi Kevin-

man watch(1)

Watch + today's nfsstat isn't smart enough to tell you deltas, or do rate conversion. However...

http://oss.sgi.com/projects/pcp/

PCP includes a gui tool which gives you rate conversion and graphical presentation with history and automatic scaling of any metrics. Plus you can do archive recording and playback, and remote access. Supported metrics include the existing server stats (but not the client-side per-server stats).

In the absence of a useful tool like PCP, some time ago I did write a patch add some of those features (rate conversion in particular) to nfsstat. The patch gives nfsstat a mode where it behaves like "vmstat 1" or "iostat 1", i.e. textually reports deltas or rates continuously in a loop with a controllable interval. I dug it up earlier today but unfortunately it won't apply anymore (due to changes for the implementation of the --sleep mode). I won't have time to work on it just now, but I can post it if anyone cares to mine it for ideas.

Signature · 3 lines
Greg Banks, P.Engineer, SGI Australian Software Group.
the brightly coloured sporks of revolution.
I don't speak for SGI.
#7 lore raw
In reply to Chuck Lever
Quoted 6 lines · Chuck Lever
Quoted 5 lines · J. Bruce Fields
Quoted 3 lines · Chuck Lever

Hi Kevin-

man watch(1)

Quoted 8 lines
Quoted 1 line

What would you watch?

For example:

watch -n3 nfsstat -c

You can also use the "-d" option to highlight the differences between the current sample and the previous sample.

He was asking for deltas; the above only gives cumulative totals.

There's no accurate one-line solution using the existing nfsstat commandline, but it should be easy to add.

--b.

Quoted 74 lines
Quoted 62 lines

--b.

Quoted 59 lines · Kevin Constantine
Quoted 44 lines

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference
between 3 seconds ago and now. Frequently I'll run a test and want to
watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

     total:        0     3476
      null:        0        0
   getattr:        0     1679
   setattr:        0        0
    lookup:        0      839
    access:        0      839
  readlink:        0        0
      read:        0        0
     write:        0        0
    create:        0        0
     mkdir:        0        0
   symlink:        0        0
     mknod:        0        0
    remove:        0        0
     rmdir:        0        0
    rename:        0        0
      link:        0        0
   readdir:        0        0
readdirplus:        0        0
    fsstat:        0      119
    fsinfo:        0        0
  pathconf:        0        0
    commit:        0        0

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 11 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" 
in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
#8 lore raw
Quoted 9 lines · J. Bruce Fields
Quoted 8 lines · Chuck Lever
Quoted 6 lines · J. Bruce Fields
Quoted 3 lines · Chuck Lever

Hi Kevin-

man watch(1)

What would you watch?

Quoted 8 lines
Quoted 6 lines

For example:

watch -n3 nfsstat -c

You can also use the "-d" option to highlight the differences between the current sample and the previous sample.

He was asking for deltas; the above only gives cumulative totals.

In that case, he might try one of the Python scripts I wrote that report NFS and RPC metrics. mountstats and nfs-iostat are included in the latest nfs-utils tarball. If they don't do exactly what is needed, it should be easy to hack them.

Quoted 79 lines

There's no accurate one-line solution using the existing nfsstat commandline, but it should be easy to add.

--b.

Quoted 74 lines
Quoted 62 lines

--b.

Quoted 59 lines · Kevin Constantine
Quoted 44 lines

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference
between 3 seconds ago and now. Frequently I'll run a test and want to
watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

    total:        0     3476
     null:        0        0
  getattr:        0     1679
  setattr:        0        0
   lookup:        0      839
   access:        0      839
 readlink:        0        0
     read:        0        0
    write:        0        0
   create:        0        0
    mkdir:        0        0
  symlink:        0        0
    mknod:        0        0
   remove:        0        0
    rmdir:        0        0
   rename:        0        0
     link:        0        0
  readdir:        0        0
readdirplus:        0        0
   fsstat:        0      119
   fsinfo:        0        0
 pathconf:        0        0
   commit:        0        0

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 11 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs"
in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 2 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com
#9 lore raw
In reply to J. Bruce Fields
Quoted 9 lines · J. Bruce Fields
Quoted 8 lines · Chuck Lever
Quoted 5 lines · J. Bruce Fields
Quoted 3 lines · Chuck Lever

Hi Kevin-

man watch(1)

What would you watch?

For example:

Quoted 9 lines
Quoted 3 lines

watch -n3 nfsstat -c

You can also use the "-d" option to highlight the differences between the current sample and the previous sample.

He was asking for deltas; the above only gives cumulative totals.

There's no accurate one-line solution using the existing nfsstat commandline, but it should be easy to add.

Something like this sort of works:
watch -n 1 'nfsstat --since /tmp/stats; nfsstat >/tmp/stats', but it feels more like a workaround than a feature.

Using watch doesn't allow you to see what happened in the past either.

Moving to a listed output format instead of the traditional nfsstat output (as seen below) makes it trivial with a simple grep to watch the stats that you really care about and ignore the rest.

-kevin

Quoted 69 lines

--b.

Quoted 67 lines
Quoted 56 lines

--b.

Quoted 54 lines · Kevin Constantine
Quoted 42 lines

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference between 3 seconds ago and now. Frequently I'll run a test and want to watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

     total:        0     3476
      null:        0        0
   getattr:        0     1679
   setattr:        0        0
    lookup:        0      839
    access:        0      839
  readlink:        0        0
      read:        0        0
     write:        0        0
    create:        0        0
     mkdir:        0        0
   symlink:        0        0
     mknod:        0        0
    remove:        0        0
     rmdir:        0        0
    rename:        0        0
      link:        0        0
   readdir:        0        0
readdirplus:        0        0
    fsstat:        0      119
    fsinfo:        0        0
  pathconf:        0        0
    commit:        0        0

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
#11 lore raw
In reply to a message not in the archive
Quoted 17 lines · Kevin Constantine
Quoted 16 lines · J. Bruce Fields
Quoted 4 lines · Chuck Lever
Quoted 3 lines · J. Bruce Fields
Quoted 2 lines · Chuck Lever

Hi Kevin-

Quoted 8 lines
Quoted 2 lines
Quoted 1 line

man watch(1)

What would you watch?

For example:

watch -n3 nfsstat -c

You can also use the "-d" option to highlight the differences between
the current sample and the previous sample.

He was asking for deltas; the above only gives cumulative totals.

Quoted 7 lines
Quoted 3 lines

There's no accurate one-line solution using the existing nfsstat commandline, but it should be easy to add.

Something like this sort of works:
watch -n 1 'nfsstat --since /tmp/stats; nfsstat >/tmp/stats', but it feels more like a workaround than a feature.

It's also slightly inaccurate, since it misses any activity that happened in the brief time between the two invocations.

Quoted 5 lines

Using watch doesn't allow you to see what happened in the past either.

Moving to a listed output format instead of the traditional nfsstat output (as seen below) makes it trivial with a simple grep to watch the stats that you really care about and ignore the rest.

I think we'd all be happy to take patches.--b.

Quoted 75 lines

-kevin

Quoted 72 lines

--b.

Quoted 70 lines
Quoted 59 lines

--b.

Quoted 57 lines · Kevin Constantine
Quoted 44 lines

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference
between 3 seconds ago and now. Frequently I'll run a test and want to
watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

     total:        0     3476
      null:        0        0
   getattr:        0     1679
   setattr:        0        0
    lookup:        0      839
    access:        0      839
  readlink:        0        0
      read:        0        0
     write:        0        0
    create:        0        0
     mkdir:        0        0
   symlink:        0        0
     mknod:        0        0
    remove:        0        0
     rmdir:        0        0
    rename:        0        0
      link:        0        0
   readdir:        0        0
readdirplus:        0        0
    fsstat:        0      119
    fsinfo:        0        0
  pathconf:        0        0
    commit:        0        0

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 11 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe 
linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
#19 lore raw
In reply to a message not in the archive
Quoted 17 lines · Kevin Constantine
Quoted 7 lines · J. Bruce Fields
Quoted 6 lines · Chuck Lever
Quoted 5 lines · J. Bruce Fields
Quoted 3 lines · Chuck Lever

Hi Kevin-

man watch(1)

What would you watch?

Quoted 8 lines
Quoted 5 lines

For example:

watch -n3 nfsstat -c

You can also use the "-d" option to highlight the differences between the current sample and the previous sample.

He was asking for deltas; the above only gives cumulative totals. There's no accurate one-line solution using the existing nfsstat commandline, but it should be easy to add.

Quoted 6 lines

Something like this sort of works:
watch -n 1 'nfsstat --since /tmp/stats; nfsstat >/tmp/stats', but it feels more like a workaround than a feature.

Using watch doesn't allow you to see what happened in the past either.

Moving to a listed output format instead of the traditional nfsstat output (as seen below) makes it trivial with a simple grep to watch the stats that you really care about and ignore the rest.

It sounds like you might find the NFS/RPC metrics reported by mountstats and nfs-iostat are what you want.

Or, you can look at the raw data in /proc/self/mountstats.

Quoted 75 lines

-kevin

Quoted 68 lines

--b.

Quoted 67 lines
Quoted 56 lines

--b.

Quoted 54 lines · Kevin Constantine
Quoted 42 lines

I'd really like to have a way to output the nfsstat counters at regular intervals (every 3 seconds) where the output is the difference between 3 seconds ago and now. Frequently I'll run a test and want to watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

    total:        0     3476
     null:        0        0
  getattr:        0     1679
  setattr:        0        0
   lookup:        0      839
   access:        0      839
 readlink:        0        0
     read:        0        0
    write:        0        0
   create:        0        0
    mkdir:        0        0
  symlink:        0        0
    mknod:        0        0
   remove:        0        0
    rmdir:        0        0
   rename:        0        0
     link:        0        0
  readdir:        0        0
readdirplus:        0        0
   fsstat:        0      119
   fsinfo:        0        0
 pathconf:        0        0
   commit:        0        0

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 2 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com
#12 lore raw
In reply to J. Bruce Fields
Quoted 23 lines · J. Bruce Fields
Quoted 14 lines · Kevin Constantine
Quoted 13 lines · J. Bruce Fields
Quoted 12 lines · Chuck Lever
Quoted 5 lines · J. Bruce Fields
Quoted 3 lines · Chuck Lever

Hi Kevin-

man watch(1)

What would you watch?

For example:

watch -n3 nfsstat -c

You can also use the "-d" option to highlight the differences between
the current sample and the previous sample.

Quoted 7 lines
Quoted 5 lines

He was asking for deltas; the above only gives cumulative totals.

There's no accurate one-line solution using the existing nfsstat commandline, but it should be easy to add.

Something like this sort of works:
watch -n 1 'nfsstat --since /tmp/stats; nfsstat >/tmp/stats', but it feels more like a workaround than a feature.

Quoted 9 lines

It's also slightly inaccurate, since it misses any activity that happened in the brief time between the two invocations.

Quoted 3 lines

Using watch doesn't allow you to see what happened in the past either.

Moving to a listed output format instead of the traditional nfsstat output (as seen below) makes it trivial with a simple grep to watch the stats that you really care about and ignore the rest.

I think we'd all be happy to take patches.--b.

Cool. I'll spend some time today then.

Quoted 73 lines
Quoted 73 lines

-kevin

Quoted 69 lines

--b.

Quoted 67 lines
Quoted 56 lines

--b.

Quoted 54 lines · Kevin Constantine
Quoted 42 lines

I'd really like to have a way to output the nfsstat counters at
regular intervals (every 3 seconds) where the output is the difference
between 3 seconds ago and now. Frequently I'll run a test and want to watch the nfs call profile throughout the course of the test.

Does something like this already exist?
Are there objections to seeing a feature like this?

I'm thinking something like:
nfsstat --sleep=1

nfs v3 call: Server Client

     total:        0     3476
      null:        0        0
   getattr:        0     1679
   setattr:        0        0
    lookup:        0      839
    access:        0      839
  readlink:        0        0
      read:        0        0
     write:        0        0
    create:        0        0
     mkdir:        0        0
   symlink:        0        0
     mknod:        0        0
    remove:        0        0
     rmdir:        0        0
    rename:        0        0
      link:        0        0
   readdir:        0        0
readdirplus:        0        0
    fsstat:        0      119
    fsinfo:        0        0
  pathconf:        0        0
    commit:        0        0

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 10 lines
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 8 lines
------------------------------------------------------------
Kevin Constantine

Systems Engineer		t: 818.460.8221
Walt Disney Animation Studios	e: kevin.constantine-P5ys19MLBK/[email protected]

Any sufficiently advanced technology is indistinguishable from magic.
    - Arthur C. Clarke
#13 lore raw
In reply to J. Bruce Fields
Quoted 29 lines · J. Bruce Fields
Quoted 17 lines · Kevin Constantine
Quoted 16 lines · J. Bruce Fields
Quoted 4 lines · Chuck Lever
Quoted 3 lines · J. Bruce Fields
Quoted 2 lines · Chuck Lever

Hi Kevin-

Quoted 8 lines
Quoted 2 lines
Quoted 1 line

man watch(1)

What would you watch?

For example:

watch -n3 nfsstat -c

You can also use the "-d" option to highlight the differences between the current sample and the previous sample.

He was asking for deltas; the above only gives cumulative totals.

Quoted 7 lines
Quoted 3 lines

There's no accurate one-line solution using the existing nfsstat commandline, but it should be easy to add.

Something like this sort of works:
watch -n 1 'nfsstat --since /tmp/stats; nfsstat >/tmp/stats', but it feels more like a workaround than a feature.

It's also slightly inaccurate, since it misses any activity that happened in the brief time between the two invocations.

Quoted 7 lines
Quoted 5 lines

Using watch doesn't allow you to see what happened in the past either.

Moving to a listed output format instead of the traditional nfsstat output (as seen below) makes it trivial with a simple grep to watch the stats that you really care about and ignore the rest.

I think we'd all be happy to take patches.--b.

Actually I would rather see the performance metrics scripts improved. These tools give a lot more information than nfsstat ever will be able to.

--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com

#14 lore raw
Quoted 3 lines · Chuck Lever

Actually I would rather see the performance metrics scripts improved. These tools give a lot more information than nfsstat ever will be able to.

Probably so, but those scripts are a bit hard to find, aren't they?

We should
- get distributions to install them by default
- write man pages?
- add references to them where possible (from the nfsstat man page, from howto's/faq's/?)

Until then, unfortunately, improvements to nfsstat are more useful, since nfsstat is the thing people are more likely to run across.

(Which might be an argument just for adding their functionality into nfsstat--that would have the one benefit that there'd be only one command name users have to know about.)

--b.

#15 lore raw
Quoted 5 lines · J. Bruce Fields
Quoted 3 lines · Chuck Lever

Actually I would rather see the performance metrics scripts improved. These tools give a lot more information than nfsstat ever will be able to.

Quoted 7 lines

Probably so, but those scripts are a bit hard to find, aren't they?

We should
- get distributions to install them by default
- write man pages?
- add references to them where possible (from the nfsstat man page, from howto's/faq's/?)

Steve promised me Red Hat would take care of this when these were added to nfs-utils last year.

Quoted 2 lines

Until then, unfortunately, improvements to nfsstat are more useful, since nfsstat is the thing people are more likely to run across.

Again, I think improving nfsstat at this point (which is merely for compatibility with Solaris) would be wasted work, if we already have what is needed in another tool. The Python tools are much more sophisticated, and it would be confusing to add their functionality to nfsstat (e.g. why can I zero the legacy stats with the -z option, but not the stats the come from /proc/self/mountstats?).

Let's spend the effort on the tools that give us deeper results.

--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com

#16 lore raw
Quoted 8 lines · Chuck Lever
Quoted 7 lines · J. Bruce Fields
Quoted 3 lines · Chuck Lever

Actually I would rather see the performance metrics scripts improved. These tools give a lot more information than nfsstat ever will be able to.

Probably so, but those scripts are a bit hard to find, aren't they?

Quoted 8 lines
Quoted 5 lines

We should
- get distributions to install them by default
- write man pages?
- add references to them where possible (from the nfsstat man page, from howto's/faq's/?)

Steve promised me Red Hat would take care of this when these were added to nfs-utils last year.

Not sure what I exactly promised (that's usually not my style 8-) ) but those scripts definitely fell off my radar...

Quoted 4 lines
Quoted 3 lines

Until then, unfortunately, improvements to nfsstat are more useful, since nfsstat is the thing people are more likely to run across.

Quoted 8 lines

Again, I think improving nfsstat at this point (which is merely for compatibility with Solaris) would be wasted work, if we already have what is needed in another tool. The Python tools are much more sophisticated, and it would be confusing to add their functionality to nfsstat (e.g. why can I zero the legacy stats with the -z option, but not the stats the come from /proc/self/mountstats?).

Let's spend the effort on the tools that give us deeper results.

I have to agree... I truly think there is a wealth of untapped information in those mountstats... Just waiting for someone to dip them out...

steved.

#18 lore raw
In reply to a message not in the archive

Hi Steve-

Quoted 15 lines · Steve Dickson
Quoted 14 lines · Chuck Lever
Quoted 12 lines · J. Bruce Fields
Quoted 3 lines · Chuck Lever

Actually I would rather see the performance metrics scripts improved. These tools give a lot more information than nfsstat ever will be able to.

Probably so, but those scripts are a bit hard to find, aren't they?

We should
- get distributions to install them by default
- write man pages?
- add references to them where possible (from the nfsstat man

     page, from howto's/faq's/?)
Quoted 4 lines
Quoted 2 lines

Steve promised me Red Hat would take care of this when these were added to nfs-utils last year.

Not sure what I exactly promised (that's usually not my style 8-) ) but those scripts definitely fell off my radar...

You promised to find an intern to compose man pages, and do the legwork needed to get these reviewed internally, and installed when nfs-utils is installed. I think that last part requires constructing a suitable Makefile.am in the nfs-utils directory where these now live, and updating the parent Makefiles to invoke it.

There was some controversy over whether nfs-utils should have a dependency on having the Python interpreter installed. Since Red Hat (for example) already uses Python for the system-config-foo utilities, this might not be too much of a problem. In any event, if we decide that Python is not appropriate, it would be simple to convert the scripts to C, or even add this functionality to nfsstat (Bruce's suggestion).

Quoted 10 lines
Quoted 10 lines
Quoted 2 lines

Until then, unfortunately, improvements to nfsstat are more useful, since nfsstat is the thing people are more likely to run across.

Again, I think improving nfsstat at this point (which is merely for compatibility with Solaris) would be wasted work, if we already have what is needed in another tool. The Python tools are much more sophisticated, and it would be confusing to add their functionality to nfsstat (e.g. why can I zero the legacy stats with the -z option, but not the stats the come from /proc/self/mountstats?).

Quoted 3 lines
Quoted 1 line

Let's spend the effort on the tools that give us deeper results.

I have to agree... I truly think there is a wealth of untapped information in those mountstats... Just waiting for someone to dip them out...

--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com