Skip to:

  1. Skip to Jira Navigation
  2. Skip to Side Navigation
  3. Skip to Main Content

PXC8.4.3 nodes crashing with "Signal SIGSEGV (Address not mapped to object) at address 0x8"

Description

Since recent upgrade to PXC8.4.3 from 8.0.35, we’ve had 2 clients reporting random but frequent crashes in their DB nodes of the form:

Both environments are 3 node clusters running on RHEL8.10

I will provide more details as I get them but is there anything we can do now to help pin this down?

Would installing the debuginfo rpms put more flesh on the bones of the stacktrace?

Environment

None

Activity

Neil Billett 
April 28, 2025 at 9:37 AM

I know this one has now been closed in favour of however I just wanted to add:

  1. The issue is easily replicable in PXC8.4.4 as well:

…and:

  1. Setting:

…on all DB nodes made no difference - crashes still occur.

This seems contrary to the findings of the OP on ticket

 

Hope its useful

Kamil Holubicki 
April 23, 2025 at 5:15 PM
(edited)

Thank you for your feedback. It seems it’s a duplicate of . Let’s continue in the other ticket.

Kamil Holubicki 
April 23, 2025 at 9:27 AM

Hi , could you please collect a coredump and check if this is the same as ?

Duplicate

Details

Assignee

Reporter

Needs QA

Affects versions

Priority

Created April 22, 2025 at 11:33 PM
Updated April 28, 2025 at 9:37 AM
Resolved April 23, 2025 at 5:18 PM