2

I'v a Linux server running under Debian Lenny with 4Go of RAM. It doesn't run a large number of stuffs:

Postfix/spamassassin (daemon mode) Bind9 KVM (one guest - 1Go of RAM for it) Every day at exactly 3:05 UTC, the server completely drop to ground floor almost all of its memory. After that, I'v more than 2 G used by buffer and never cleaned up (unless I manually tell the kernel to drop the cache).

I'v searched the web a lot and, at the begining, I though this was due to NFS buffer usage. I do backup over an NFS share drive using gzip/tar and the backup occured at 3:05.

However, I'm now in a very strange situation because I moved the backup task at 1:40 (it completes in 2 mins) and I still drop all the RAM at 3:05.

In my logs, nothing particular, except that at 03:05:01, cron open a session as root and immediately close it at 03:05:02 without doing anything. Of course, cron has been restarted and I checked the timing of the tasks - again, nothing particular.

Any idea why this happens? Or, any idea about how to track what's using all those buffers?

Thanks for your help,

Henry-Nicolas Tourneur
  • 21
  • 2
  • Are you running low on memory? What does `free -m` report? – Steven Monday Apr 30 '11 at 14:48
  • What kind of cron job is it that runs at 3:05? Could you check your root's crontab? – Axel Knauf Apr 30 '11 at 18:25
  • to Steven:free -m will report ~2G used for buffer, 100M free, 100M for cache and the rest is in use by apps – Henry-Nicolas Tourneur Apr 30 '11 at 21:06
  • to Axel: it used to be a perl script that performs backup on an NFS share drive using gzip. Now the same script runs at 1:40 but still the same issue at 3:05. – Henry-Nicolas Tourneur Apr 30 '11 at 21:07
  • Here you can see what happens: http://img13.imageshack.us/i/snapshot2u.png/
    At the begining of the free space increase, the backup script is runned. At the end of the space (return to 100M free), it is 03:05. Some times ago, that script used to run at 03:05 but now it runs at 01:40. The script is mounting an NFS drive, gzipping folder of the server on the NFS share then it umount the share and exit. This is strange, in the first place, I though the script was the cause of the memory usage but now it's more like it's somehow forcing some internal clean-up.     – Henry-Nicolas Tourneur Apr 30 '11 at 21:16
  • Linux kernel is using buffer and cache memory to speed-up tasks (when repeated). How it handle it is internal. As long as your system and services do not suffer from the exchange of data between these 2 cache, I'd not worry. It could be that the mapped block IO (in the buffer) are just flushed by your backup script, and so the kernel can use more memory for caching pages (in the cache). – Huygens Apr 25 '12 at 09:29

0 Answers0