As discussed before first step is too

edit
/etc/cron.daily/1hup_apache_logs

Change it too look like


then edit
/etc/rc.d/init.d/httpd_app_init

find reload-grc

In pico ctrl+w

change


to:



The reason people have the errors happend at random times is because everytime u add / edit / delete a domain it does a "reload-grc" which is causing it. Now why it is causing it im not sure, have a client who experiencing it and spent last 2 hours figuring out a fix...

Enjoy!

Cool find ameen

I updated the link in my checklist.

I have allready apply the official ensim httpdmon.so fix.
It is needed to make and this?

Is this Long Lost Child error related to the "child pid 1693 exit signal Segmentation fault (11)" error by any chance? I've see the child error before when typing "ps aux", and it was often near a "defunct " line, which looks to be the same as the Segfault error. Any help in figuring out what is causing these segfault errors would be MOST appreciated. It is crashing my machine.

we all seem to be in the dark here, i'm looking for the same fix for a while now, at least here there a new step "killproc httpd -USR1" to killproc httpd -HUP" in the init which the techno staff dude at RS never mantion, Bravo! and Amen to u, Ameen.

can't believe that I thought 3.1 was better then 3.0, i have both and i never have anything major with 3.0 so for all of you guy who thing about upgrade...if it an't broken don't touch it!

just give the child an old-fashioned spanking, ground'm for a week and make sure he/she never leaves home without your permission ever again.
yup.

just give the child an old-fashioned spanking, ground'm for a week and make sure he/she never leaves home without your permission ever again.

Can you explain more detail ?

I have the same problems.

I think 3.1.1 is worse that 3.0 and 3.1.0

I want to buy a new one, which version do you guy recommend ?

The latest and the greatest, version 3.1.72

i can see one small problem with implementing this, if you have resellers that keeps add/edit/deleting sites , and apache restarts rather then reload gracefully......your other virtual sites is going to be disrupted pretty badly.

however if you don't , this is a good fix.
I have implemented only half of this, done the apache_logs change to reload but skipping the changing of reload-grc from USR1 to HUP..

and things appears to be working fine with a average of 5 million hits a day.

is this because war is near and the world is maybe at stake here that finally someone post in the topic, took exactly 4 months, good that we are not under any sort of DEADline here....killing of corpse.

but seriously the problem just started again in the last few days, so for over 90 days my server was ok, the apache was keeping crashing one in a while but lately it has start and twice, even with sim2.2 it hasn't restart, and that after over 12 to 14 repeated try from sim to restart the devil (like this better then deamon).

i'm not what to try next, other then asking Rackshack to set an other server with of course a different version of ensim, which mine is 3.1.0-25 on a celeron 1.3, 512, 40hd , at $1 setup fee I supposed i could simply get an other machine, but who will pay for all the effort that i put in setting this one properly, i suppose me.

ensim aren't very useful about this bug, they knew but didn't do much to fix it, at least not on the version of ensim that i used.

and at last i could have try to migrate my ensim to 3.1.1 or 3.1.3 but that again doesn't garanti that it would solve it.

so this is it we are near the break of an horible event and i'm here @#!@#!@#!@#!@#!ing about one of my server, what a pitty.

cu

- jy

Anybody has a follow up on this problem ??? (didn't find any on the forum)

The only response i got from tech support is to check my files for bad rights. They say it's a memory access problem.

Have u guys got any other leads to go on ?

the only thing that can fix that problem is a restore, order one from RS they will put a new drive on you server, which should have a lastest ensim, for me it is 3.1.8-2. this fix my problem for good. just ask them to keep the old drive for backup.

good luck

It's already update to the latest (rpm updated) the box his only 1 week old so a restore won't do the trick and i really doubt it would change anything since the errors in the log files are present even before the rpm update.

I managed to pull the restart-grc -HUP trick so apache can restart itself but i still get the annoying segmentation fault in the error and apache always goes down for ~1min and goes up again.

crap algo, sad to hear that, i though that the newer ensim would do the trick, all i could say is put a trouble ticket about it and hope for the best, worst thing ask them to switch server, btw what is it, celeron, p4 and what is the ensim version who are currently using. when i got mine back i didn't touch it at all, other then security mesure, it is set like out of the box.

btw I'm from MTL like you cool, pm me if you which, like to know more about fellow from my town sometime.

take care, and sincerly good luck, i know how pain it is i have been thru the same, yuk

It's a Celeron 1.7 (should be 1.3 but phpsysinfo sais 1.7 i won't argu lol) Can't remember the original version of Ensim when i first got it but now it's 3.5.10-12.

I did a trouble ticket on RS but they sent me the same crap like the others talking about defective ram and so on but it's really a Ensim issue and they ain't steping up to the plate to correct or acknowledge the problem.

I prefer the Dell P3 800 i got in Mtl has no panel but stable like rock.. with 30 clients it stills purs like a kitten and has a 0.0 load average.. uptime 63 days and counting.

Great How-To ameen. Thanks.

Apache on my Plesk 7 server dies after every few hours with following message in error logs:-

SIGUSR1 received

I tried to fix with the help of this thread, but my plesk server does not have these files:-

/etc/cron.daily/1hup_apache_logs

/etc/rc.d/init.d/httpd_app_init

How could I fix the problem on Plesk server?

Thanks
UKAGG