hp-health (Debian kernel backport) - segfault

[Exploitation67]

Hi

 My server is:

• ProLiant DL380 G7

Firmware:

• SPP 2016.10

 System:

• Debian 8.6 (Jessie)
• Linux Kernel 4.8.15-2~bpo8+2 ( backport )
• hp-health ver.  10.40-1815.49

Syndrom:

When hp-health is started I got hpasmlited segfault error:

hpasmlited -f /dev/hpilo

 [1289761.888536] x86/PAT: hpasmlited:29216 map pfn expected mapping type uncached-minus for [mem 0xdf7fe000-0xdf7fefff], got write-back
[1289761.888620] show_signal_msg: 14 callbacks suppressed
[1289761.888621] hpasmlited[29216]: segfault at 0 ip 000000000040605b sp 00007ffe96425e30 error 4 in hpasmlited[400000+3a000]

Will HP take a look and create a new version of package ?

Best Regards.

Exploit

[GregDavis]

We don't directly test or support Debian, but we do test and support certain Debian clones.  I'll pass your information to my team and see if there are any concerns about the impact of this problem on future builds of the Debian clones we do support.

[Jimmy Vance]

This will be fixed in the nex release of the hp-health package

[Exploitation67]

Hi,

Great news.

Do you know when this new version will be released ?

Regards,

Exploit

[dmautz]

I'm seeing a very similar crash on CentOS 7.2.1511 w/ Kernel 4.6.4, running hp-health-10.50-1826.40.rhel7.x86_64.  Is the root cause the same as the issue reported above?  Thank you!

Mar 17 11:06:25 x01 hpasmlited[3273381]: StartIpmiRmqThread
Mar 17 11:06:25 x01 hpasmlited[3273381]: StartIpmiRmqThread: Thread created (TID -1987787008)
Mar 17 11:06:25 x01 hpasmlited[3273381]: IpmiRmqThread: Entry
Mar 17 11:06:25 x01 hpasmlited[3273381]: ehpsmb_parse_SMBIOS: SMBIOSInitTable was not successful.
Mar 17 11:06:25 x01 kernel: [2751957.060306] x86/PAT: hpasmlited:3273381 map pfn expected mapping type uncached-minus for [mem 0x79173000-0x79175fff], got write-back
Mar 17 11:06:25 x01 kernel: [2751957.060397] hpasmlited[3273381]: segfault at 0 ip 0000000000421d7b sp 00007ffec8fe4fc0 error 4 in hpasmlited[400000+39000]

[rpronk]

When can we expect an update of the packages with an included fix for this problem?

I would love to be able to use it with the Jessie backports kernel. It's quite an important toolset for our servers.

[velin]

Any news on the new version of hp-health with that fix ?

[Allibtech]

Same problem with last version of hp-health (10.60) and debian stretch 9.1.

dmesg

[111014.496270] x86/PAT: hpasmlited:32232 map pfn expected mapping type uncached-minus for [mem 0x788d0000-0x788d1fff], got write-back

systemctl status hp-health
● hp-health.service - LSB: hp System Health Monitor and Command line Utility Package.
   Loaded: loaded (/etc/init.d/hp-health; generated; vendor preset: enabled)
   Active: active (exited) since Fri 2017-10-20 15:11:53 CEST; 18min ago
     Docs: man:systemd-sysv-generator(8)
  Process: 19801 ExecStop=/etc/init.d/hp-health stop (code=exited, status=0/SUCCESS)
  Process: 19848 ExecStart=/etc/init.d/hp-health start (code=exited, status=0/SUCCESS)
    Tasks: 0 (limit: 7372)
   CGroup: /system.slice/hp-health.service

Oct 20 15:11:53 vms8 hpasmlited[19909]: IpmiRmqThread: Entry
Oct 20 15:11:53 vms8 hpasmlited[19909]: (SendRecv_CHIF) CpqCiSend: error=4, "Bad name for the target connection object.".
Oct 20 15:11:53 vms8 hpasmlited[19909]: (SendRecv_CHIF) CpqCiSend: error=4, "Bad name for the target connection object.".
Oct 20 15:11:53 vms8 hp-health[19848]: /etc/init.d/hp-health: line 665: /usr/lib/systemd/scripts/hp-snmp-agents.sh: No such file or directory
Oct 20 15:11:53 vms8 hp-health[19848]:
Oct 20 15:11:53 vms8 systemd[1]: Started LSB: hp System Health Monitor and Command line Utility Package..
Oct 20 15:11:53 vms8 hpasmlited[19909]: ehpsmb_parse_SMBIOS: SMBIOSInitTable was not successful.
Oct 20 15:11:53 vms8 hpasmlited[19909]: Not able to initialize BIOS interfaces!
Oct 20 15:11:58 vms8 hpasmlited[19909]: StopIpmiRmqThread
Oct 20 15:11:58 vms8 hpasmlited[19909]: StopIpmiRmqThread: Thread stopped

Eric