Server 10 Disk hda7 (/tmp) is at 100%

My site is down.

/tmp was full of temporary mod_zip files. We've fixed it and your site is working now.
Sorry for the caused inconveniences.

Thank You

All my websites on server 10 are down again.
Looks like you have disk space at maximum.
Server Load 2.05 (2 cpus) <<<<<<<<<----------
Memory Used 24.6 %
Swap 0.938 %
Disk hda2 (/) 13 %
Disk hda1 (/boot) 32 %
Disk hda8 (/home) 72 %
Disk hda7 (/tmp) 100 % <<<<<<<<<----------
Disk hda6 (/usr) 91 % <<<<<<<<<----------
Disk hda5 (/var) 74 %

You guys REALLY need to watch this alot closer and take action BEFORE disaster strikes!!!!!!

Thank you for superquick response!
Awesome.

Please see the last sentence of the note of Oct 10.

Server 10 /usr is once again at 97% and rising.

DON'T YOU GUYS WATCH ANY OF THIS?
Or do you wait for catastrophe?

You really need to tighten up your monitoring and automatic PREVENTIVE actions.

I admit, you are usually very responsive ONCE something is brought to your attenetion, but it should NOT have to wait until then.

We do watch of this. You just noticed that earlier then we did.

I have been getting complaints from my clients about "slow response, timeouts, poor performance"

For the past several weeks, I have been checking up on server10 stats. What I almost ALWAYS find is that Server load is extremely high. Server10 has two CPUs. A few minutes ago server load was at 16.17 and hovered in the range of 10 to 18 for approximately FIFTEEN minutes. Then it came down to 1.80 for a few minutes, then back up to 5.28 and hovered at 4 - 8.

I have randomly checked stats during the day. RARELY do I find it below two.

Are there some users or some applications that are misbehaving? Or should Hostony consider adding CPU's to server10 or replace server10?

Thanks.

This is strange. I tried to check it too and didn't see it more than 5.0.
I'll monitor it more.

Status as of 3pm Eastern US Time Notice the Server Load

apache (1.3.28 (Unix)) up
exim (exim-4.24-30_cpanel_stmpcontrol_antivirus_rewrite_mailman2) up
ftpd failed <=================
imap failed
interchange up
mysql (4.0.15-standard) failed <=================
syslogd up
webmail up
Server Load 111.86 (4 cpus) <=================
Memory Used 26.3 %
Swap 0.925 %
Disk sda3 (/) 64 %
Disk sda1 (/boot) 15 %
Disk sdb1 (/home) 40 %

I've been having the same problems for several weeks, and it is server load and seem this morning, it was just a complete failure as I can't log in to anything

GMT 4:54pm up 5 days

This is a cut from #uptime
So server is up for 5 days. Still can't see the problem with it.

I am only talking about server 10, of course. I don't know what's going on with other people's servers. At random times during the day, I can catch server 10 well above 2 CPU's for many minutes at a time. I really do not want to have to start collecting statistical data about this, but if I get any more complaints from clients about slow and dead response, I will have to do something to convince you that SOMETHING is happening on server 10, at least some of the time ... too much of the time.

Right NOW:
apache (1.3.28 (Unix)) up
exim (exim-4.24-30_cpanel_stmpcontrol_antivirus_rewrite_mailman2) up
ftpd up
imap up
interchange up
mysql (4.0.15-standard) up
syslogd up
webmail up
Server Load 11.42 (2 cpus)
Memory Used 47 %
Swap 0.743 %
Disk hda2 (/) 13 %
Disk hda1 (/boot) 32 %
Disk hda8 (/home) 82 %
Disk hda7 (/tmp) 5 %
Disk hda6 (/usr) 64 %
Disk hda5 (/var) 66 %

Thanks.
JimC

As far as server10 was one of the shy servers we had. I don't refuse that something may happen on the server at certain times because of some cronjobs run by users or something else. For last week we have a person dedicated just for monitoring server load and performance of all servers because we had problems with 2 servers being DDOSed and did not noticed any problems with server 10.
We need soem extra info from your users whn they notice a problem. At least a tracert to the server or error they get at the browser as well as exact time. it should help us to determine a problem if any.

Of course, TRACERT will tell you if the user has a slow HOP, but if the slow performance is because the server CPU is being killed, TRACERT is irreleveant.

Much of this morning:

apache (1.3.28 (Unix)) up
exim (exim-4.24-30_cpanel_stmpcontrol_antivirus_rewrite_mailman2) up
ftpd up
imap up
interchange up
mysql (4.0.15-standard) up
syslogd up
webmail up
Server Load 9.77 (2 cpus) <<<=========================much of this morning
Memory Used 42.6 %
Swap 0.802 %
Disk hda2 (/) 22 %
Disk hda1 (/boot) 32 %
Disk hda8 (/home) 86 % <<<=========================creeping up all weekend
Disk hda7 (/tmp) 4 %
Disk hda6 (/usr) 73 %
Disk hda5 (/var) 72 %

Support Ticket 2911

As of Nov 24, 2003, 6PM EST:
Disk hda8 (/home) 90 %
Disk hda5 (/var) 83 %

/dev/hda8 43G 37G 4.6G 89% /home
/dev/hda5 3.8G 2.3G 1.3G 63% /var

there are 4.6 Gigabytes free on /home so it is not problem.