Chkservd on the fritz?

[DigiCrime]

Whats up with it this morning? I wake up this morning and find 400 system messages, all disk space messages "Drive Space Warning on server.someexample.com" no big deal maybe it was a fluke but whatever is wrong with that program, its only sending the drive space warning messages, duplicates, over and over, and 40+ servers, im getting roughly 12-15 of these per min I went in and restarted chkservd on ten systems to see if there was any effect but appears to be none at all. Ive looked at a handful of systems, rebooted a couple, doesn't seem to have any effect either. Checked the log to see if chkservd was failing

Aug 2 09:06:43 earth sshd(pam_unix)[10801]: session opened for user root by (uid=0)
Aug 2 09:07:06 earth chkservd: chkservd shutdown succeeded
Aug 2 09:07:06 earth chkservd: chkservd startup succeeded

Where would I go from here? All systems are using the same Cpanel Release build, Mixed operating systems from Redhat 7.x to Fedora, 30 machines have the same hardware, same drives same, nics, ram pretty much everything so I dunno

[thomas]

Same problem here.

[rover]

Having same problem here also

[dwayrynen]

Whats up with it this morning? I wake up this morning and find 400 system messages, all disk space messages "Drive Space Warning on server.someexample.com" no big deal maybe it was a fluke but whatever is wrong with that

cpanel pushed out an update for /usr/sbin/chkservd that broke how disk space is checked.

Prior to the update the code that determined if a disk space check was going to go through was this:


$dskchk++;
if ($dskchk == 1) {
diskcheck();
}
if ($dskchk == 289) {
$dskchk = 0;
}

Now it is this:

if ($dskchk % 300 == 0 || $dskchk == 0) {
diskcheck();
}


Both are similar - previously disk space checks were only running every 290 * 500 seconds - or about every 40 hours or so...


Now it's being checked every 500 seconds....

You can either clear up the disk space (-:, or change that second snippet of code to be
something like:


if ($dskchk++ % 300 == 0 || $dskchk == 0) {
$dskchk++;
diskcheck();
}

You have to increment the counter in two places because I think the intent is to check disk space on start up and then every 300 iterations.

Personally, I think the disk space monitoring should be configurable, both on time between checks, and the amount of space left before warnings are issued - the defaults don't cover everyone's needs very well and should be tunable with some predefined defaults.

Darin

[cpdan]

Hello,

If anyone else has this problem (a couple of you had tickets already) have a look here for a temporary fix and subscribe yourself to the CC list to be notified of any changes to it.

http://bugzilla.cpanel.net/show_bug.cgi?id=973

Thanks

[DigiCrime]

ah yea thanks dan I was getting ready to update my thread for everyone else who was having this trouble.

Be nice if it was configurable to allow you to set an amount of used space on a specific partition/drive, we get them even when a drive is at 80% thats not very active. just used space

[pirania1]

I guess I had related problem.
The only difference was TONs of emails regarding urchin and urchin scheduler failing to start, even though they were started properly.

[iski4kix]

The bug report fix is not working for me any other ideas?

[mike25]

Using the method above I have been able to disable disk checking. This will stop those annoying emails but it is also not what I would prefer to do. Instead I would like to change the % at which these emails are sent out. Does anyone know where the code is that determines this?