iSeries Services not Starting correctly or passing Port Test

jean_driscoll

jean_driscoll

VIP Member
Our Development server, that has been up and running for years and has had no recent changes, all of a sudden does not pass the port test. I'm getting the following errors in the server logs for port test. has anyone seen this before? I've been waiting all day for a response from PS.

642 Thu Oct 14 09:40:49 2004 dbdrvsql.c441
OS40002001 - Error Type: OTHER enum: 12 - QBFCPRCED API internal error!

642 Thu Oct 14 09:40:49 2004 jdb_drvm.c695
JDB9900164 - Failed to connect to System - B7333

642 Thu Oct 14 09:40:49 2004 jtp_cm.c283
JDB9909003 - Could not init connect.

642 Thu Oct 14 09:40:49 2004 jtp_tm.c960
JDB9909100 - Get connect info failed: Transaction ID =

642 Thu Oct 14 09:40:49 2004 jdb_rq1.c1942
JDB3100013 - Failed to get connectinfo

In another log I see this:
648 Thu Oct 14 10:00:34 2004 ipcpub.c3135
process 648 <B7334SYS/JDENET_N> registered in entry 0

648 Thu Oct 14 10:00:41 2004 netflow.c426
648 (pid) in routeOutgoingQueue(), unable to send message msgType 11 to host=<172.16.2.23>
 

Attachments

  • 80552-Jde-jdedebug after porttest.doc
    96 KB · Views: 518
Hi,
There is a doc oti-99-0037 in the KG on 12 possible errors in the Porttest. Maybe someone changed the CFGTCP - lowercase name for the host, when it should be uppercase?
...DLK
 
One reason for the error is if your development server is connecting to the production server for system tables and the database code levels are different between the two. SI14585 is the latest level of the QBFC/QXDA code. If it is on one server only, you can have problems.

- Clark Scholten
 
thanks for the suggestion. I do have that ptf on order, but not applied to either system. These two systems have been working correctly for years. We have applied all of the same patches that were applied to this development server to the production server, so we should be fine on that front.
 
Something perhaps to check. I've seen instances where one of the IBM SQLPKG's was missing or corrupt causing the OS40002001 issue, but the error message was a little different. Have you first verified that you haven't lost one or more, and if not, perhaps restoring them from the last backup where you know the system was running correctly.

Regards,
 
Along the lines of that, you can delete the QZDAPKG package and have it recreate itself. The other two, you need to get from IBM or a backup.
 
What are the names of the other two packages? I have the qZDApkg sqlpkg. I'm guessing I would have to stop the QUSRWRK sbs in order to delete this package. Then would just starting the sbs back up recreate it?
 
Don't quote me, and hopefully someone with access to a iSeries will give a definitive answer, but I believe the two others are QSQXDPKG and QSQLPKG2.

Regards,
 
thanks. I've got those 2 packages. I will try deleting the qzdapkg. It will recreate itself when I restart the subsystem, correct?
 
I deleted that package and it did get recreated when the subsystem started back up. unfortunately I still can't get port test to work. here is the message I am receiving: Socket error on recv: A connection with a remote socket was reset by that soc
ket.
Initialization of environment failed - rcode = 0

In the joblog is this message: Cannot resolve to object QSOC. Type and Subtype X'0401' Authority X'000 which is a library that I don't have on the system, never have. I don't have the library on my production system either and port test is working fine over there.

I'm thinking somehow I turned on some function like interoperability by mistake, but I don't remember doing anything last week to this iSeries, at all.

It's so frustrating, all that the PS tech is doing is having me go through the port test paper over and over. He's not listening.
 
QSOC appears to be OptiConnect related, we don't have that library on our system either. :confused:
 
How are you connecting to the data sources on your other AS/400? Are you using JDBNET or a virtual LAN connection if you have a system with LPARs?
 
I think I've been getting that message all along, too. But it's the only error in my joblog. My issue is that although users can sign onto the system, they cannot submit jobs. I've removed all members from the Printqueue table (this is development) so I know the problem is not related to that.

Services do start, but they have errors in the logs that say:
388 Mon Oct 18 16:41:55 2004 ipcpub.c3135
process 388 <B7334SYS/JDENET_N> registered in entry 0

388 Mon Oct 18 16:42:02 2004 netflow.c426
388 (pid) in routeOutgoingQueue(), unable to send message msgType 11 to host=<172.16.2.23>

388 Mon Oct 18 17:03:37 2004 netsig.c513
Net program ended, pid = 388, signal = 6

388 Mon Oct 18 17:03:37 2004 ipcpub.c3622
API ipcSawZombieProcV1 : process 388 set to Zombie in entry 0

388 Mon Oct 18 17:03:37 2004 netsig.c513
Net program ended, pid = 388, signal = 0

388 Mon Oct 18 17:03:37 2004 ipcpub.c3643
API ipcSawZombieProcV1 : process 388 not found in process SHM


and this:
ct 18 16:41:56 2004 cleanup.c139
Initializing Environment:DV7333 . . .

390 Mon Oct 18 16:42:02 2004 dbdrvsql.c441
OS40002001 - Error Type: OTHER enum: 12 - QBFCPRCED API internal error!

390 Mon Oct 18 16:42:02 2004 jdb_drvm.c695
JDB9900164 - Failed to connect to System - B7333

Again, that QBFCPRCED program is not found on either server and Production is working fine. the PS guy keeps asking me to load the program by applying a PTF that I've already told him is already applied (si04904).
 
Hey Jean, are there any other errors? The QSOC error is standard functionality when using DBDR as it attempts to look for Opticonnect even if the product is not installed on the iSeries. It then should time out and revert to TCP/IP communications. However, the system should still come up and function, even after the error logs show this error. I'm just thinking that there could be something else going on. Did you perchance try restoring those other two SQL packages from a time when the system was working properly?

Regards,
 
Just another thought, though I'm sure it's not it...is the EDRSQL server running?

Regards,
 
thanks, yes, that was the first thing I checked, as I've had that issue before.
 
We had the exact same error before, beside making sure the EDRSQL server is started, did you check the connection status, on the NETSTAT 'Work with TCP/IP connection status' find the remote system you want to connect and the remote port of as-edrsql, see if they are in "Established" state. Our problem before was someone swap our host table entry around (using SNA instead of TCP). Hope this help
 

Similar threads

Getting Error while calling BSSV from E9 Web.
Replies
2
Views
4K
BossDBK
Back
Top