Homepage › Clovertech Forums › Read Only Archives › Cloverleaf › Cloverleaf › Recover_56
- This topic has 12 replies, 6 voices, and was last updated 14 years, 11 months ago by Richard Hart.
-
CreatorTopic
-
April 14, 2008 at 9:35 am #49965David HarrisonParticipant
Coverleaf 5.6 on Solaris 10 I
-
CreatorTopic
-
AuthorReplies
-
-
April 14, 2008 at 1:23 pm #64289John HamiltonParticipant
It look correct. This kind of error is indicative of an error in the procedure.
Not sure if you copied your scripts from the previous version or these came with your distribution.
When I came here and move to their new server I had a very similar error. I had to modify the recovery proc there was a syntax error in it.
-
April 14, 2008 at 1:30 pm #64290David HarrisonParticipant
I downloaded these scripts from the Technical Bulletins section of this site – posted by Rob Abbot. I’ve just found a syntax error in the downloaded recover_56.tcl. Line 330 incr send_cn should be incr send_cnt.
This hasn’t fixed my problem though.
-
April 14, 2008 at 1:54 pm #64291David HarrisonParticipant
There is a line in the proc which should get the handle of the outbound message but it returns a null value: Code:keylget args OBMSGID my_mh ;# Get handle of outbound message
-
April 14, 2008 at 1:58 pm #64292Charlie BursellParticipant
Why did you install the resend proc? If you checked the resend OB messages box there would be no place to put this proc. Problems such as this should be reported to Heathvision Support (972)361-3004. Please contact Support. Send them a copy of your procs and your NetConfig. Tell them I told you to contact them and they will send me the files and I will take a look
Charlie
-
April 14, 2008 at 2:24 pm #64293David HarrisonParticipant
I checked the Reply Generation box. I need to modify the check_ack proc in order to modify and resend the message dependent upon the received ack and the message type. Can you give me an email address for Healthvision support as I am in the UK.
Thanks,
Dave
-
April 14, 2008 at 2:35 pm #64294Charlie BursellParticipant
If you look at the Report you will see there is a bug with using the old metod that will be fixed the end of this month. I acn’t tell if that is your problem or not without seeing your configuration.
-
April 14, 2008 at 2:41 pm #64295David HarrisonParticipant
I’ve done as you’ve suggested. Thanks,
Dave
-
May 8, 2009 at 6:15 pm #64296Michael HertelParticipantQuote:
There is a line in the proc which should get the handle of the outbound message but it returns a null value:
Code:
keylget args OBMSGID my_mh ;# Get handle of outbound message
Alka is having this issue.
Was there any resolution?
-
May 11, 2009 at 11:03 am #64297David HarrisonParticipant
It’s working as far as I’m concerned. On the outbound, for Inbound replies timeout I checked Resend OB Message, and for tps inbound reply I’m using check_ack.
-
November 3, 2009 at 7:59 pm #64298Carter HarrisonParticipant
I too have run into this issue. We are using the check_ack scripts that were provided by Healthvision in the 5.6 release notes. 99% of the time everything works great, but occasionally the check_ack script is called and the value in OBMSGID is null. I just sent in a detailed report to support but have not heard anything back yet. Was a solution to this problem ever found?
-
November 5, 2009 at 5:11 am #64299Richard HartParticipant
Cloverleaf 5.6 AIX
Possibly related. My code in prewrite, send_data_ok and sms_ib_reply.
I’m sending multiple types of message from different threads, out bound from one Cloverleaf thread to a web service. For one thread, I need to pass the original message on for further processing.
In my code I had – for the ‘special’ thread (see below). I’ve tried a few variations, even creating a new message and using data from the OBMSGID, but this always seems to crash – sometimes 3 messages, sometimes 500 messages!
It is WIP at the moment.
if {[keylget args OBMSGID myObMsgId]} {
set myMsgId [msgcopy $myObMsgId]
msgmetaset $myMsgId DESTCONN {} SOURCECONN $HciConnName
lappend myDispList “CONTINUE $myMsgId”
}
This crashes inconsistently, but with the same issue …
[tps :tps :INFO/2:gds_tsit_ih_mlt_sib:11/05/2009 08:27:07] [midNULL] Unbind message0 from TPS Interp
[tps :tps :INFO/1:gds_tsit_ih_mlt_sib:11/05/2009 08:27:07] tds.string = gds_xlate_canonical {MSGID message2} {CONTEXT sms_ib_reply} {ARGS {}} {MODE run} {VERSION 3.0}
….
Encode OK
[msg :Msg :DBUG/0:gds_tsit_ih_mlt_sib:11/05/2009 08:27:07] [0.0.10] msgFree 0x30002528
gds_xlate_canonical {KILL message2} {CONTINUE message0}
[diag:leak:DBUG/0:gds_tsit_ih_mlt_sib:11/05/2009 08:27:07] diag md alloc 0x211ef2c8
[tps :tps :INFO/2:gds_tsit_ih_mlt_sib:11/05/2009 08:27:07] [midNULL] Unbind !^^
-
November 5, 2009 at 8:37 am #64300Richard HartParticipant
Probably not related!
I tried a few things without success and then removed the extra procedure calls – for debug printing and translation – and then all seems OK – for 1500 message anyway!
-
-
AuthorReplies
- The forum ‘Cloverleaf’ is closed to new topics and replies.