Peter van Dijk wrote:
>On Sat, Apr 15, 2000 at 06:09:45AM -0400, Curtis Generous wrote:
>> "Beware that NFS, async filesystems, and softupdates filesystems may
>> discard files that were not safely written to disk before an outage"
>>
>> Besides this failure mode (which syslog also suffers from), and there
>> any other restrictions to logging over NFS with multilog?
>
>No, altho performance might be disappointing - you will have to try that
>for yourself.
It seems to me that the right balance between the two would be to
multilog to local files, and once a log is rotated out of current, move it
to your central NFS. You won't have a perfectly up-to-date snapshot on your
NFS disk, but it'll fix the performance issue.
I haven't poked, but as I read http://cr.yp.to/daemontools/multilog.html
you could probably automate this using !processor.
Alternately, have cron job move all the rotated multilog files once an
{hour,day,week,whatever}.
--
gowen -- Greg Owen -- [EMAIL PROTECTED]
