[exim-dev] [Bug 1601] New: hundrets of segfaults : Fedora 21…

Góra strony
Delete this message
Reply to this message
Autor: Resellerdesktop Admins
Data:  
Dla: exim-dev
Temat: [exim-dev] [Bug 1601] New: hundrets of segfaults : Fedora 21 | 3.17.7 i686
------- You are receiving this mail because: -------
You are on the CC list for the bug.

http://bugs.exim.org/show_bug.cgi?id=1601
           Summary: hundrets of segfaults : Fedora 21 | 3.17.7 i686
           Product: Exim
           Version: 4.84
          Platform: x86
        OS/Version: Linux
            Status: NEW
          Severity: bug
          Priority: medium
         Component: Unfiled
        AssignedTo: nigel@???
        ReportedBy: customercare@???
                CC: exim-dev@???



Note: A similar Bugreport is open in the Fedora Bugtracker.

At leats since the 16.2. 2015 one server in a cluster of many gets this
segfaults:

Mar 21 09:55:55 s36 kernel: exim[21626]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+
11b000]
Mar 21 09:55:55 s36 kernel: exim[21626]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]

Decoding the ip causing it, we get a multicast ip ( the other way around , it's
an unassigned ip ). The system does not have multicast enabled, nor does it get
packets from that ip address.

20 other servers are runing with the same setup, the same package, the same
network traffic, the same {add more sames here} .. fine, without segfaults in
exim.

I checked the yum.log for updates, but it did not relate any exim updates to
the startdate of the segfaults. It came out of the blue.

You will notce the same segfault over and over again. It's the untampered (
besides the config of course ) fedora exim version out of the official repo.

Mar 20 21:57:50 s36 kernel: [589489.682104] exim[28781]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 20 21:57:50 s36 kernel: exim[28781]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 20 22:20:27 s36 kernel: [590846.284077] exim[15987]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 20 22:20:27 s36 kernel: exim[15987]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 20 22:53:31 s36 kernel: [592830.473130] exim[18697]: segfault at 0 ip
b76a58e7 sp bfbe6020 error 6 in exim[b7600000+11b000]
Mar 20 22:53:31 s36 kernel: exim[18697]: segfault at 0 ip b76a58e7 sp bfbe6020
error 6 in exim[b7600000+11b000]
Mar 21 01:42:54 s36 kernel: [602993.332914] exim[3768]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 01:42:54 s36 kernel: exim[3768]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 01:52:14 s36 kernel: [603553.997435] exim[17239]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 01:52:14 s36 kernel: exim[17239]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 01:54:53 s36 kernel: exim[19449]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 01:54:53 s36 kernel: [603712.969029] exim[19449]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 02:29:43 s36 kernel: [605802.288511] exim[17130]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 02:29:43 s36 kernel: exim[17130]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 02:36:55 s36 kernel: [606234.931952] exim[22581]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 02:36:55 s36 kernel: exim[22581]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 03:27:04 s36 kernel: [609244.248269] exim[22737]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 03:27:04 s36 kernel: exim[22737]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 03:52:07 s36 kernel: [610746.716401] exim[8235]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 03:52:07 s36 kernel: exim[8235]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 05:22:51 s36 kernel: [616190.416454] exim[31673]: segfault at 0 ip
b76a58e7 sp bfbe6020 error 6 in exim[b7600000+11b000]
Mar 21 05:22:51 s36 kernel: exim[31673]: segfault at 0 ip b76a58e7 sp bfbe6020
error 6 in exim[b7600000+11b000]
Mar 21 07:43:29 s36 kernel: [624628.723615] exim[25618]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 07:43:29 s36 kernel: exim[25618]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 08:09:22 s36 kernel: [626181.343220] exim[14691]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 08:09:22 s36 kernel: exim[14691]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 08:33:15 s36 kernel: [627615.042879] exim[4416]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 08:33:15 s36 kernel: exim[4416]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 08:37:34 s36 kernel: [627873.605902] exim[7709]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 08:37:34 s36 kernel: exim[7709]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 08:41:46 s36 kernel: [628125.281821] exim[11055]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 08:41:46 s36 kernel: exim[11055]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 09:28:49 s36 kernel: [630948.613335] exim[25061]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 09:28:49 s36 kernel: exim[25061]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]
Mar 21 09:55:55 s36 kernel: [632574.811667] exim[21626]: segfault at 0 ip
b76a58e7 sp bfbe6910 error 6 in exim[b7600000+11b000]
Mar 21 09:55:55 s36 kernel: exim[21626]: segfault at 0 ip b76a58e7 sp bfbe6910
error 6 in exim[b7600000+11b000]


Sidenote:

The server has a high amount of emails to transport. A general strace of all
running exims to get the segfault, isn't a good or doable way. If you can
somehow be more specific what strace shall log i.e. -e
open,stats,close,write,read etc etc, that may help to ease the debug ioload a
bit.


--
Configure bugmail: http://bugs.exim.org/userprefs.cgi?tab=email