In <list-9640009@2rosenthals.com>, on 05/11/24
at 11:02 AM, "Massimo S." <ecs-isp@2rosenthals.com> said:
Hi Massimo,
I've reviewed the testlog and the yum output uploaded to the ticket.
There's nothing obvious in the data that would explain the bind errors.
FWIW, against my wishes, Massimo has chosen to edit the testlog and remove
items that he believes are sensitive data and that, without sufficient
technical knowledge, he believes have nothing to do with the issue.
Hopefully, for his sake he has guessed right. If not, I will unassign
myself from the ticket and let him resolve the bind hang on his own or
with help from others.
Massimo, I recommend you review the testlog and correct the reported PATH
typo.
The server is basically running eCS 2.2b with some updated drivers and
some really old drivers. This means the server is running a kernel and
PSD combo that have many SMP defects which have been fixed in newer kernel
and PSD releases. Whether this is the reason for the named hangs is
unknown at this time.
The server is running a circa 2018 of the mmlem.os2 network driver.
Whether or not this is contributing to the hangs is unknown at this time.
Next time bind hangs, please capture a process dump and send me a link to
the 7z archive. Try to remember to capture the process dump before
attempting to kill named.
tilbake listen