Open Server Error: 16238.10.0: srv_senddone(): Invalid result type(1010)

[S Rosenblum]

We are running Adaptive Server Enterprise/12.5.2/EBF 11799/P/Linux
Intel/Enterprise Linux/ase1252/1831/32-bit and have been for some time
without issue. Our night database backup failed last night with the
following error:
Open Server Error: 16238.10.0: srv_senddone(): Invalid result type(1010)
for event(2)

Never happened before. If have not been able to location information on
this error message or actions I need to take.

Thank you

[A. M.]

Post by S Rosenblum
We are running Adaptive Server Enterprise/12.5.2/EBF 11799/P/Linux
Intel/Enterprise Linux/ase1252/1831/32-bit and have been for some time
without issue. Our night database backup failed last night with the
Open Server Error: 16238.10.0: srv_senddone(): Invalid result type(1010)
for event(2)
Never happened before. If have not been able to location information on
this error message or actions I need to take.

I don't know what it is either but what I would do is
bounce both the ASE and backup servers and start monitoring
for it to happen again. When you run backups, take snapshots
of what's happening on ASE as well. Its unlikely to be
anything else running on ASE so the snapshots are more
to disprove that it is anything else.

Were there any other messages in the server logs around
the same time?

-am © 2007

[S Rosenblum]

Thank you for responding.

When this even occurred, we were getting an uncharacteristic spike in
activity leading to overall memory issues. The errorlog contains memory
related errors (below)

Error: 709, Severity: 17, State: 1
server There is insufficient system memory to continue login process
for spid 505.

We have had the memory related situation before but have never had any
impact on the backup processes. Maybe this is because we have never had
them occur concurrent to a backup procedure.

I do plan to cycle Sybase when we have a maintenance window. I was
hoping I could find a description of the error prior to cycling given
that it may not come back up....

Any insights are welcome. Thanks.

Post by A. M.

Post by S Rosenblum
We are running Adaptive Server Enterprise/12.5.2/EBF 11799/P/Linux
Intel/Enterprise Linux/ase1252/1831/32-bit and have been for some time
without issue. Our night database backup failed last night with the
Open Server Error: 16238.10.0: srv_senddone(): Invalid result type(1010)
for event(2)
Never happened before. If have not been able to location information on
this error message or actions I need to take.

I don't know what it is either but what I would do is
bounce both the ASE and backup servers and start monitoring
for it to happen again. When you run backups, take snapshots
of what's happening on ASE as well. Its unlikely to be
anything else running on ASE so the snapshots are more
to disprove that it is anything else.
Were there any other messages in the server logs around
the same time?
-am © 2007

[A. M.]

Post by S Rosenblum
When this even occurred, we were getting an uncharacteristic spike in
activity leading to overall memory issues. The errorlog contains memory
related errors (below)
Error: 709, Severity: 17, State: 1
server There is insufficient system memory to continue login process
for spid 505.

Ah! Well that makes things a little more interesting.

Usually, if all user connections are allocated, you would
get another message saying something along the lines of
"can't connect". However, error 709 suggests that there
was at least one (if not more) user connections available
but that other factors prevented a successful creation of
a connection. In this case its a memory configuration issue.
Here's the write up for the error -
http://infocenter.sybase.com/help/topic/com.sybase.39996_1250/html/svrtsg/svrtsg194.htm
(scroll down to the 709 write up. Its at the bottom).

Basically, you'll need to review your server's configuration
settings for total memory. You might also need to check the
proc cache. See the 'action' part of the write up. Unfortunately,
any config change will require a bounce (but not for the
reasons/assumptions I originally made).

Post by S Rosenblum
We have had the memory related situation before but have never had any
impact on the backup processes. Maybe this is because we have never had
them occur concurrent to a backup procedure.

Yes, that appears to be the case. It would have been nice
to know what else was happening/running on the server at
the time. It looks now as if it was proc cache related.
If its related to an errant process, you might want to
set up some sort of monitoring to see if you can track
it down.

Post by S Rosenblum
I do plan to cycle Sybase when we have a maintenance window. I was
hoping I could find a description of the error prior to cycling given
that it may not come back up....

Its hard to say. If a new process/routine was added recently,
and its the one causing proc cache issues, it may start
occurring more frequently. Then again, it could just be
a one-off because there were, in fact, a large number of
connections. If that was the case, then it just implies
a proc cache sizing issue.

-am © 2007