Jeremy Chadwick wrote:
On Fri, Dec 07, 2007 at 08:44:37AM +0100, Patrick van Iersel wrote:
From: Eugene Grosbein [EMAIL PROTECTED]
See http://www.freebsd.org/cgi/query-pr.cgi?pr=75855
It seems the regression from RELENG_4 is still here.
There are so many regressions in performance that I
From: Eugene Grosbein [EMAIL PROTECTED]
Date: Fri, 7 Dec 2007 14:48:47 +0700
To: Patrick van Iersel [EMAIL PROTECTED]
Cc: freebsd-stable@FreeBSD.org
Subject: Re: Extreme load with local password db lookups
On Fri, Dec 07, 2007 at 08:44:37AM +0100, Patrick van Iersel wrote:
See http
I would like to point out that in FreeBSD 7.x there is a daemon called nscd.
I
believe it was created exactly for this purpose (speeding up name lookups by
caching them).
Is it true that 4.x has nearly O(n) lookup speed while later versions
has O(n^2) method? Why 4.x does not need caching
On Friday 07 December 2007, Jeremy Chadwick wrote:
On Fri, Dec 07, 2007 at 08:44:37AM +0100, Patrick van Iersel wrote:
From: Eugene Grosbein [EMAIL PROTECTED]
See http://www.freebsd.org/cgi/query-pr.cgi?pr=75855
It seems the regression from RELENG_4 is still here.
There are so many
On Friday 07 December 2007, Eugene Grosbein wrote:
I would like to point out that in FreeBSD 7.x there is a daemon called
nscd. I believe it was created exactly for this purpose (speeding up name
lookups by caching them).
Is it true that 4.x has nearly O(n) lookup speed while later
Hi,
Here's the situation. We have an web/ftp server with around 74000 users
defined in the local unix password database. On 4.9-stable which it is
running now, there is no noticable load when lookups are done (logins via
ftp, ~user lookups from apache etc.). We want to migrate this system to
On Thu, Dec 06, 2007 at 11:37:50AM +0100, Patrick van Iersel wrote:
Here's the situation. We have an web/ftp server with around 74000 users
defined in the local unix password database. On 4.9-stable which it is
running now, there is no noticable load when lookups are done (logins via
ftp,
On Thu, Dec 06, 2007 at 06:22:07PM +0700, Eugene Grosbein wrote:
On Thu, Dec 06, 2007 at 11:37:50AM +0100, Patrick van Iersel wrote:
Here's the situation. We have an web/ftp server with around 74000 users
defined in the local unix password database. On 4.9-stable which it is
running now,
Patrick van Iersel wrote:
However on 6-STABLE (FreeBSD 6.3-PRERELEASE #0: Wed Dec 5 13:35:05 CET
2007) these same lookups cause very high load and things slow down to a
crawl.
Hi,
How do you know it's the passwd lookups that cause the problems? (I'm
not saying they're not...)
A common way to debug is to isolate the affected systemcalls, by using
tools like ktrace or strace.
Strace can also record a timestamp, you can see how long it takes to
complete a specific systemcall
# strace -r -f serverproc
Do a single login and then examine the results.
happy debugging.
[EMAIL PROTECTED], Patrick van Iersel
[EMAIL PROTECTED],freebsd-stable@freebsd.org
Subject: Re: Extreme load with local password db lookups
A common way to debug is to isolate the affected systemcalls, by using
tools like ktrace or strace.
Strace can also record a timestamp, you can see how
Jeremy Chadwick wrote:
Here's the situation. We have an web/ftp server with around 74000 users
defined in the local unix password database. On 4.9-stable which it is
running now, there is no noticable load when lookups are done (logins via
ftp, ~user lookups from apache etc.). We want
On Fri, Dec 07, 2007 at 08:44:37AM +0100, Patrick van Iersel wrote:
See http://www.freebsd.org/cgi/query-pr.cgi?pr=75855
It seems the regression from RELENG_4 is still here.
There are so many regressions in performance that I still
prefer to invest time to patch bsd.ports.mk to support
From: Eugene Grosbein [EMAIL PROTECTED]
Organization: SVZServ
Date: Fri, 07 Dec 2007 12:40:42 +0700
To: Jeremy Chadwick [EMAIL PROTECTED]
Cc: Patrick van Iersel [EMAIL PROTECTED], freebsd-stable@FreeBSD.org
Subject: Re: Extreme load with local password db lookups
Jeremy Chadwick wrote
On Fri, Dec 07, 2007 at 08:44:37AM +0100, Patrick van Iersel wrote:
From: Eugene Grosbein [EMAIL PROTECTED]
See http://www.freebsd.org/cgi/query-pr.cgi?pr=75855
It seems the regression from RELENG_4 is still here.
There are so many regressions in performance that I still
prefer to
15 matches
Mail list logo