[Date Prev][Date Next]
[Chronological]
[Thread]
[Top]
Re: Tweaking "threads"
- To: openldap <openldap-software@OpenLDAP.org>
- Subject: Re: Tweaking "threads"
- From: Ben Beuchler <insyte@gmail.com>
- Date: Thu, 1 Sep 2005 10:49:34 -0500
- Content-disposition: inline
- Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=Xb5UwzP6SMBUyau/WTQu5NdVw6++2kC16pf58FFOY2FAl35eTPqAt6xhbvVoz0RowMpEBXtoNVJg4P8Tt1+hO4gC3nqFCXtQ3kmyZlM73OIqinVwdA8znSBzAGs3qaq0DuRjldh9b76M8jjaRmrmK1UXh1p1znGgn12HBE7X3vo=
- In-reply-to: <479b70ed05083122426518045@mail.gmail.com>
- References: <479b70ed05083122426518045@mail.gmail.com>
On 9/1/05, Ben Beuchler <insyte@gmail.com> wrote:
> I'm suspicious that individual LDAP sessions may be sticking around a
> lot longer than they should, as I am seeing a *lot* more "BIND" log
> entries than I am "UNBIND". For example:
>
> swizzle:/var/log# grep ' BIND' slapd.log|wc -l
> 142308
> swizzle:/var/log# grep ' UNBIND' slapd.log|wc -l
> 36520
This may have been a red herring. About an hour ago I set the
"idletimeout" to 30.
This is what the numbers look like on the logs created since then:
swizzle:/var/log# grep ' BIND' working.log|wc -l
12237
swizzle:/var/log# grep ' UNBIND' working.log|wc -l
974
Still pretty disparate.
So looking a little more closely at the logs, I decided to count
'ACCEPT' and 'closed' entries for the whole log, not BIND and UNBIND:
swizzle:/var/log# grep ACCEPT slapd.log|wc -l
16139
swizzle:/var/log# grep closed slapd.log|wc -l
16226
Additionally, I set "idletimeout 30" about 3 hours ago. We haven't
even hit the busiest part of the day and I'm already running at 57
slapd threads. Is that also a red herring? Should I be concerned
about needing that many threads?
Thanks!
-Ben