WHM Crahses RHE

[Myacen]

Every minute:

Jan 16 08:22:04 pluto kernel: application bug: whostmgrd(19521) has SIGCHLD set to SIG_IGN but calls wait().
Jan 16 08:22:04 pluto kernel: (see the NOTES section of 'man 2 wait'). Workaround activated.

an 16 08:23:06 pluto kernel: application bug: whostmgrd(20008) has SIGCHLD set to SIG_IGN but calls wait().
Jan 16 08:23:06 pluto kernel: (see the NOTES section of 'man 2 wait'). Workaround activated.


I am on the latest edge. It eventually causes the kernel to panic.

[Myacen]

I should say this was copying an account btw.

[Angel78]

From LogWatch : Rhe 3.0 with latest release

5 Time(s): (see the NOTES section of 'man 2 wait'). Workaround activated.
1 Time(s): application bug: whostmgrd(17915) has SIGCHLD set to SIG_IGN but calls wait().
1 Time(s): application bug: whostmgrd(5896) has SIGCHLD set to SIG_IGN but calls wait().
1 Time(s): application bug: whostmgrd(6120) has SIGCHLD set to SIG_IGN but calls wait().
1 Time(s): application bug: whostmgrd(6327) has SIGCHLD set to SIG_IGN but calls wait().
1 Time(s): application bug: whostmgrd(6823) has SIGCHLD set to SIG_IGN but calls wait().


btw, care to share your pkgskiplist (for up2date)?

[Myacen]

I think it's something with the copy account thats where we got kernel panics

[thedavid]

You guys been having issues with that too? I sent a note to cpanel support regarding these (and a ton of other) errors that are occuring only on the rhe box. They responded by telling me that they don't support kernels when I asked if they were planning on addressing the issue...

-David

[cpanelnick]

This won't actually cause any problems other then generate that annoying message.

[Myacen]

Sorry to say nick, but the kernel is panicking on our servers and others with RHE & cpanel. Not sure what the issue is however even /scripts/fixcommonproblems locks up one of our servers.

[MattDr2]

I noticed a similar problem... when loading WHM on a machine the server crashed. All I did was load WHM. D:

But that was on RH9.

Regards,
Matt

[sgemma]

Same issue here... server runs fine, underloaded even, here are the stats:

Dual Xeon 2.0 GHZ
1.5 GB RAM

Here are the kernel logs right before the machine crashes:

Feb 13 11:11:01 w10 kernel: application bug: whostmgrd(29386) has SIGCHLD set to SIG_IGN but calls wait().
Feb 13 11:11:01 w10 kernel: (see the NOTES section of 'man 2 wait'). Workaround activated.

Anybody find a solution?

Perhaps the RedHat 2.4.20-28.9smp kernel has some issues with WHM?

[nat]

Same issue here... server runs fine, underloaded even, here are the stats:

2.4 GHz Pentinum 4
1.5 GB RAM
WHM 8.5.4 cPanel 8.5.5-E65
RedHat Enterprise 3 - WHM X v2.1.1

Here are the kernel logs right before the machine crashes:

Feb 16 04:45:03 server kernel: application bug: whostmgrd(26244) has SIGCHLD set to SIG_IGN but calls wait().
Feb 16 04:45:03 server kernel: (see the NOTES section of 'man 2 wait'). Workaround activated.

Anybody find a solution?

[nat]

Just crashed again. last thing in logs:
Feb 16 13:33:46 server kernel: application bug: whostmgrd(20011) has SIGCHLD set to SIG_IGN but calls wait().
Feb 16 13:33:46 server kernel: (see the NOTES section of 'man 2 wait'). Workaround activated.


This is what cpanel sent to me:

--

This is not something to worry about as found in the cPanel CHANGELOG.

this is really annoying since we are going to be forced to call
wait in some circumstances even when SIGCHLD is set to IGNORE
because of broken kernels. The tradeoff is the error message which
doesn't affect anything but makes people go crazy when they see it.
However its better that we don't create zombie processes, so for
compatibility between diffrent linux versions it might just be
best to live with this warning message as it doesn't do anything
but make people ask questions.

--

Well its funny how the server just keeps on stopping to work after getting that error. It's nothing to worry about.-Whatever. Why not just fix it?