Progress_Guy
Member
Hi folks,
We've recently upgraded servers and moved our databases and character clients to the new servers (HP-UX11, 32-bit). The character clients used to connect shared memory, but, we now connect them client server - we need to do this as we are now using a clustering DR environment and the CHUI apps may be on a different server than the dbases - in this case they are all on the same serevr and we are not doing anything like a remote proshut.
We had a UNIX HP parameter set a little low and it has limited the number of telnet sessions on the box. As a result we are having to kill of Progress users using promon and found that when killing a user the user briefly disappeared from the user list and then reappered again.
Upon checking the dbase log we saw the following :-
10:44:30 SRV 1: Logout usernum 45, userid Andyg, on ux-cn1 tj. (739)
10:44:37 SHUT 5: User 73 disconnect initiated
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
73
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
77
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
85
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
102
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
109
10:44:44 SRV 1: Login usernum 47, userid Andyg, on ux-cn1 tj. (742)
I've seen loads of postings on here and progress.com and have a rough idea where the problem lies (transactions till open, lock on user table, no client proecss existing, etc.), but, would appreciate your thoughs and any possible solution/fix.
This is already with Progress Tech. Support, but, I'd appreciate a little help as it's quite urgent.
Many Thanks,
Dave.
We've recently upgraded servers and moved our databases and character clients to the new servers (HP-UX11, 32-bit). The character clients used to connect shared memory, but, we now connect them client server - we need to do this as we are now using a clustering DR environment and the CHUI apps may be on a different server than the dbases - in this case they are all on the same serevr and we are not doing anything like a remote proshut.
We had a UNIX HP parameter set a little low and it has limited the number of telnet sessions on the box. As a result we are having to kill of Progress users using promon and found that when killing a user the user briefly disappeared from the user list and then reappered again.
Upon checking the dbase log we saw the following :-
10:44:30 SRV 1: Logout usernum 45, userid Andyg, on ux-cn1 tj. (739)
10:44:37 SHUT 5: User 73 disconnect initiated
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
73
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
77
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
85
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
102
10:44:37 BROKER 0: brsigsome:Notifying srvr 3, conn 3, to terminate remote user
109
10:44:44 SRV 1: Login usernum 47, userid Andyg, on ux-cn1 tj. (742)
I've seen loads of postings on here and progress.com and have a rough idea where the problem lies (transactions till open, lock on user table, no client proecss existing, etc.), but, would appreciate your thoughs and any possible solution/fix.
This is already with Progress Tech. Support, but, I'd appreciate a little help as it's quite urgent.
Many Thanks,
Dave.
