|
news about PUB400
|
This page contains latest news to be read by all. Discussions to be conducted in the forum page.
2026-10-13
Again a (too long) downtime. What happened the last days?
I prepared to move to a neat Power10 box (sponsored by IBM, thanks!)
While doing so, i created a SAVE21 to VTL, then restored this tape
to the new system connected to a new SAN with external virtualized SSD.
seemed to be OK in the first time, but after 2 days i noticed several issues due to damaged objects...
so this must be severe. First suspected a system crash, but in the end, it maybe is the SAN and
the combination with Power10 and external virtualized disks. This needs to be verified in the next days.
In the meantime, PUB400 runs on our Power11 and SAN. But this box does not have the proper licensing for all
the developers. So time is running out. Do not worry about messages concerning 5770WDS. It will be
solved as soon as i fixed the storage issue.
Then again a planned downtime is to be expected...
2025-11-04
Phew! This was a longer unplanned downtime. Sorry for that.
What happened? A system object was defunct, and the regular IPL
for loading updates tried to update this object, which failed.
The system ended up in manual state, and i was on travel...
When being back i tried to cleanup, which consists of:
full save, reclaim storage (test all objects)
clean up, reboot again...
as this system runs on a rather small box, but has a high amount
of objects, output and job queues, this process needs long time.
As i needed some sleep somewhen, next day came, and i found some
other objects which stopped the system save to run. So further
cleaning up...
in the end, it was successful but needed 2-3 days. will check
with IBM, so did a flash copy and created a second LPAR to
analzye deeper.
Please report any issues by mail. Hopefully i will get
a faster machine next year.
Stay tuned, soon PUB400 has its 25th anniversary!
2025-08-20
this is the first time i needed to lock out some users and their respective email domain.
they used Database access to scan all users objects and produced a lot of error messages (due to security settings)
please keep in mind this is a public system, you should not disturb other people!
2025-06-11
Currently there is an issue with spools (might be mixed between users).
To work this out with IBM plase send me an email with your WRKSPLF output and notes on which spool is not displayed correctly or may not be owned by you.
first steps to fix this is to run a RCLSPLSTG (since some hours) and a RCLSTG (which needs downtime).
So if you cannot reach PUB400, simply wait :-)
mail to hs(at) rzkh.de