My experience with ESX is that when VM is freshly restarted, disk speed is
close to bare metal, but longer the VM is up the worse it gets. In my case
it went from 60MB/s to cca 1.5MB/s.
b.
On 18 November 2011 00:15, Igor Cicimov wrote:
> Hmmm that's a pretty bad disk read I would expect this t
Seconded:)
b.
On 10 November 2011 02:03, Igor Cicimov wrote:
> I would first look into PHP if you are running any php app...
>
> On Wed, Nov 9, 2011 at 11:13 PM, Bostjan Skufca wrote:
>>
>> Any allow,deny directives in use? That coupled with hostname (i.e.
>> local
Any allow,deny directives in use? That coupled with hostname (i.e.
localhost) turns reverse lookups back on.
What if you telnet to you apache and issue the request manually, does
it wait for 10 seconds before spitting out the content? Did you strace
the server process?
b.
On 9 November 2011 12:
Does your server support IPv6? Try removing the ::1/128 part.
b.
On 31 October 2011 15:28, Nick Riesland wrote:
> Hello,
>
> I'm getting this error in the Event Viewer when I try to start Apache2.2
>
> *+* *System*
> *-* *Provider*
>[ *Name*] Apache Service
> *-* *EventID* 3299
>
This means that your httpd process is stuck waiting for file descriptor
number 7.
You should now look at what this file descriptor is. "lsof" command is
useful here: "lsof -p PID" where PID is the PID number of the stuck process.
In the fourth column look for number 7 followed by either r, w or u
Can't reproduce it with the same versions/arch. Although I have noticed
those urls lately on one of servers I am taking care of.
The only difference is: i'm using *:80 as Listen address, not IP:80.
b.
On 24 October 2011 18:02, Svenne Krap wrote:
> Hi.
>
> I am currently investigating a misbe
This is the extract from source file server/mpm/prefork/prefork.c:
/* Limit on the total --- clients will be locked out if more servers than
* this are needed. It is intended solely to keep the server from c
mount -o noatime,nodiratime,remount /path/to/mount/point
b.
On 28 May 2011 20:13, Geoff Millikan wrote:
> > I will mod fstab and reboot one server at a time...
>
> I thought there was a way of dropping atime in real time without a reboot
> so you can test performance differences without reboo
most of what you are requesting… I will see about
> making them open to the world for a couple days for you to take a peek at…
>
>
>
> Thanks for your reply, its very much appriciated…
>
>
>
> Have a great weekend!
>
>
>
> Rob Morin
>
> Systems Administrato
Hi Rob,
couple of thoughts below, please inform me if I make wrong deductions
somewhere, tnx.
1. Your static content is served from elsewhere, you are only talking about
application servers
2. Your average response time is 200ms, and given 4cores per server and
excluding delays you are currently
On 9 May 2011 15:44, Matus UHLAR - fantomas wrote:
> So that was bad way to simulate "apachectl stop" just because of the above.
> I think with 2.2 it even is not true, because apache2 has own way to
> configure shutdown timeouts, the GracefulShutdownTimeout directive.
Yes, timeout is configurab
Just tried it on one of my servers, I get a ratio of 0,7% in ssl vs non-ssl
ab benchmark, without keepalive. With keepalive it gets to 1/3.
However, I noticed something else a year or so ago, never figured out the
cause.
I can't get apache to work faster than 11k req/s, no matter how many
concurre
>
> if a server is killed (SIGKILL) during a "large" static file transfer, then
>> the client is not notified by his browser that file has not been completely
>> downloaded. On Win it just says it is not a valid Win32 application or
>> corrupted or sth.
>> Now I know this is not a general problem a
Hi all,
if a server is killed (SIGKILL) during a "large" static file transfer, then
the client is not notified by his browser that file has not been completely
downloaded. On Win it just says it is not a valid Win32 application or
corrupted or sth.
Now I know this is not a general problem and a gr
Hi everybody!
I have noticed that Apache (2.2.14/linux) is causing a lot of page faults
while serving any request, be it static file or some application server
stuff.
Correlation goes like this:
- benchmark is started with 4 ab processes with 32 clients each.
- benchmark results at around 2k requ
15 matches
Mail list logo