Hello Wazuh Community
We have a Test Installation with Agent v4.2.4 running on Solaris SPARC 11.4 SRU38. After Updating Solaris to SRU41 and XSCF to 3110. There came up some KERNEL Panics, which were followed up by a Hardware reboot.
We decided then to remove the wazuh Agent, und then with every uninstallaion inside a Zone and the Global Zone, the KERNEL Panic came up again.
Is there anybody having the same issues?
The Panic Message was the following:
panic[cpu35]/thread=6400b2bf4ec0: BAD TRAP: type=31 rp=2a10710a860 addr=0 mmu_fsr=0 occurred in module "unix" due to a NULL pointer dereference
wazuh-logcollect: trap type = 0x31
pid=1175, pc=0x1002a824, sp=0x2a10710a101, tstate=0x80001601, context=0xa7d
g1-g7: 2, 208d9878, 208d9800, ca, 8, a7d, 6400b2bf4ec0
Support tells us, that the cause of the panic is clear. It looks like the current directory of this process is null.
Solaris Support also asked on which SRU the Binarys (especially v4.2.4) at the Webpage are builded. Could some answer this question?
Many thanks in Advance
Andy
Hello Luis,
Firstly, I would ask Oracle for the patch that solves this problem, that would be the best approach.
Secondly, If no luck with Oracle support, try this: mkdir -p /backup/bin; mv -f /var/ossec/bin/* /backup/bin and then, restart the system, the service should fail to start, then proceed to remove the wazuh-agent package.
Let us know if this helped!
John.-