| 1 |
Paul Hartman написав(ла): |
| 2 |
> On Wed, Jun 10, 2009 at 1:24 PM, Alexander |
| 3 |
> Pilipovsky<alexander.pilipovsky@×××××.com> wrote: |
| 4 |
> |
| 5 |
>> 2009/6/10 Justin <justin@×××××××××.net> |
| 6 |
>> |
| 7 |
>>> Alexander Pilipovsky schrieb: |
| 8 |
>>> .... |
| 9 |
>>> |
| 10 |
>>> You have something big in /var. Check /var/tmp/portage or /var/log. |
| 11 |
>>> |
| 12 |
>>> Alos test sys-fs/ncdu or similar tools. |
| 13 |
>>> |
| 14 |
>>> |
| 15 |
>> Thanks, it's was /var/log/messages that had 11 GB and was not opened by any |
| 16 |
>> editor. |
| 17 |
>> |
| 18 |
> |
| 19 |
> I had a similar problem, 100% cpu usage caused by a daemon that did |
| 20 |
> not respond well when network connection was lost. My |
| 21 |
> /var/log/messages grew over 60GB in a few hours with the same message |
| 22 |
> repeated tens of millions of times. I changed my logger from syslog-ng |
| 23 |
> to metalog, which suppresses duplicates, and now it's not a problem. |
| 24 |
> |
| 25 |
Thanks, I'm going to install metalog :) |
| 26 |
|
| 27 |
-- |
| 28 |
Alexander Pilipovsky aka Engraver |
Archives