SYSTEM ERROR: strent request for more than 32K. (893)

[aries_rusli]

Dear all,

Hi my name is Aries, I hope I can find the solution from this forum.

I have multivolume database ( 1 bi, 11 data extend ) around 11Gb and using Progress 8.3C and o/s windows 2000 server.

I got my database off suddenly and can't start after that ( even server has been restarted), I saw in lg file, it showed the error

SYSTEM ERROR: strent request for more than 32K. (893)
SYSTEM ERROR: Releasing regular latch. latchId: 4 (5028)
SYSTEM ERROR: Releasing regular latch. latchId: 1 (5028)
User 62 died holding 2 shared memory locks. (2522)
Stopped. (2520)
Stopped. (2520)

Does anyone know, what caused this problem ? Any solution ?
I don't do anything in the programming since we use the existing software application.
I never got this error before ( from year 2000 )

Thanks for your help.

 

[jakke]

...
SYSTEM ERROR: strent request for more than 32K. (893)
...

Hi. Dont know if it will help you but we also got this strent error in our db-log few weeks ago. Knowledge base entry told us the issue can appear when running an app for longer time. (it had to do something with a counter which is incrementing internally during for-each's or do-while's)
Took a while to identify the issue but we found it to be 2 programs which created ton's of open handle's (inside some huge for each).
(The error was always preceeded by a few memory increases)

I see your situation is a bit different but...

Grts

 

[Casper]

I got my database off suddenly and can't start after that

Well you probably have the database already running again, but if not:

Is it not even possible to start the database with a simple proserve statement like:
proserve <pathtodatabase><databasename>

What errors you get when starting the database?

Regards,

Casper.

 

[senetd]

Symptoms:
Client session dies
Client server connection to a database
SYSTEM ERROR: Strent request for more than 32K
Error message only in log file of the Database
Stack trace from _mprosrv reads: strentd strent qrGetQuery qrOpen

Cause:
Too many queries are left open on the remote server, because they are not closed properly on the client side. Then the mproserve process crashes.

Fixes:
Find out in the code the queries that are left opened and manage to close them. Using the new Debugger (9.1D06 or later version) with Dynamic Object Tracking can help to identify the offending queries.


From progress KB; Dejan Šenet