Operating System - HP-UX
1748246 Members
3781 Online
108760 Solutions
New Discussion юеВ

Oracle 10G - sgslufread: Hard error on read (OS errors 32 and 232)

 
Alzhy
Honored Contributor

Oracle 10G - sgslufread: Hard error on read (OS errors 32 and 232)

Anyone experiencing this on 11.11 and Oracle 10G? WOuld anyone please share their SR Cases raised with Oracle? Is there a known HP-UX 11.11 patch to address this?


sgslufread: Hard error on read, OS error = 232
sgslufread: Hard error on read, OS error = 32


Also, what do error 232 and 32 mean?
A dump of sterrcat yields:

32 Too many links
232 Software caused connection abort

Hakuna Matata.
3 REPLIES 3
Yogeeraj_1
Honored Contributor

Re: Oracle 10G - sgslufread: Hard error on read (OS errors 32 and 232)

Hi nelson,

we are running on a similar environment, where were these errors reported?

We can try to verify our systems here.

kind regards
yogeeraj
No person was ever honoured for what he received. Honour has been the reward for what he gave (clavin coolidge)
Sherry Clark
New Member

Re: Oracle 10G - sgslufread: Hard error on read (OS errors 32 and 232)

I am getting the same error with oracle 10.2.0.2 on our production HP running 11.23 during copy of data from remote databases to local database. I have opened an SR with oracle (6150859.994). We had our hp support look into it and they responded:

The error code 232 indicates "Connection reset by peer" and the 32
indicates "Broken pipe". Basically sqlplus opens connection (socket) to
the Oracle database, or more specifically the Oracle TNS listener.
Something caused the Oracle TNS listener to reset the connection to
sqlplus.

We are showing no errors in any of the oracle logs (alert, listener, sqlnet).
dik pater
New Member

Re: Oracle 10G - sgslufread: Hard error on read (OS errors 32 and 232)

I ran into the same problem.
Just migrated from 8.1.7.4 on vms to 10.20.20 on hpux itanium 11.23. We extract data with parallel query option, in 8.1.7.4 when we had an integrity violation we often ran into ORA-03113, it seems that we now have the same problem, but now we have no ORA error but a sqllufread, it takes a very very long time to receive this error by the way.
I will test this again with no parallization on, and will enter more information to this forum.