[Bug 1842020] Re: ceph patch as of 8/29 segfaults all bluestore osds

[Harry Coin]

The above data is from a run in an eoan VM on the same processor that hosts the osd's in the baremetal layer (same bug in both cases, but I reverted the baremetal layer to the state prior to the 'upgrade' to restore ceph osd function).
Here's the dpkg -l you asked for.

Here's cat /proc/cpuinfo on the VM
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 15
model name      : Intel Celeron_4x0 (Conroe/Merom Class Core 2)
stepping        : 3
microcode       : 0x1
cpu MHz         : 2327.284
cache size      : 16384 KB
physical id     : 0
siblings        : 1
core id         : 0
cpu cores       : 1
apicid          : 0
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 10
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx lm constant_tsc rep_good nopl cpuid tsc_known_freq pni vmx ssse3 cx16 x2apic tsc_deadline_timer hypervisor lahf_lm cpuid_fault pti tpr_shadow vnmi flexpriority tsc_adjust arat arch_capabilities
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs
bogomips        : 4654.56
clflush size    : 64
cache_alignment : 64
address sizes   : 40 bits physical, 48 bits virtual
power management:

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6
model           : 15
model name      : Intel Celeron_4x0 (Conroe/Merom Class Core 2)
stepping        : 3
microcode       : 0x1
cpu MHz         : 2327.284
cache size      : 16384 KB
physical id     : 1
siblings        : 1
core id         : 0
cpu cores       : 1
apicid          : 1
initial apicid  : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 10
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx lm constant_tsc rep_good nopl cpuid tsc_known_freq pni vmx ssse3 cx16 x2apic tsc_deadline_timer hypervisor lahf_lm cpuid_fault pti tpr_shadow vnmi flexpriority tsc_adjust arat arch_capabilities
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs
bogomips        : 4654.56
clflush size    : 64
cache_alignment : 64
address sizes   : 40 bits physical, 48 bits virtual
power management:


** Attachment added: "list of packages installed on system with crashes."
   https://bugs.launchpad.net/ubuntu/+source/ceph/+bug/1842020/+attachment/5285813/+files/dpkg-l.txt

-- 
You received this bug notification because you are a member of Ubuntu
OpenStack, which is subscribed to ceph in Ubuntu.
https://bugs.launchpad.net/bugs/1842020

Title:
  ceph patch as of 8/29 segfaults all bluestore osds

Status in ceph package in Ubuntu:
  New

Bug description:
  The ceph patch on eoan distributed 8/29 crashes all bluestore OSD's
  and is unusable at least on some systems.  Failed on an old dual Xeon
  E5345 box.

  Easy test.  Run:

  /usr/bin/ceph-bluestore-tool

  On working systems it reports a help message.  On the latest eoan release distributed by canonical it
  turns in to an illegal instruction process kill, just after reading /proc/<...>/auxv

  looks like something to do with vsock issues.

  strace -k -y /usr/bin/ceph-bluestore-tool 
  is instructive.

  Notice valgrind against that program reports thousands of memory allocation issues.
  Reverting the system to the snapshot before the apt upgrade restores full operations.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/ceph/+bug/1842020/+subscriptions