Forum Replies Created
-
AuthorReplies
-
I haven’t tried it, but you can probably use the same images folder and update the 6.2 jar. Make sure you make a backup copy first! 🙂
Steve
We had the same issue when we upgraded from 5.7 to 6.1.2. We had to add ‘-usetls 0’ to get it working (came from Infor):
smtp::sendmessage $messageToSend
-usetls 0
-header
-header
-header
-servers $serverName
Hope this helps.
Steve
I updated my jar file from 6.1.2.1 with the images folder from the jar Farhan provided for 6.1.2.
No need to restart the host server. The new entry is available once the file is saved.
Steve
Is the Cloverleaf icon showing up in the taskbar? Do you use dual monitors in your office? I have occasionally had programs show up in the taskbar, but they are not on the screen. You can press the windows key + left arrow a couple of times to move it to the main screen.
Steve
Charlie,
At the time this was done, we didn’t realize the obvious. Since it has worked for us for so many years, we just never looked to change. I found a previous post of yours that recommended using ‘/usr/bin/env tcl’ and that’s the direction that we’re headed.
Russ – You are correct. That was our goal many years ago. ‘/usr/bin/env hcitcl’ also works.
Thank you both for the replies.
Steve
Correction. If I comment out the package require for clsqlite and use the symbolic link, a package require for sqlite will fail.
$HCIROOT/tcl/lib doesn’t contain a sqlite directory in 6.1.
At this point, until Infor supplies a patch, my best option is to change my scripts to use ‘/usr/bin/env tcl’ instead of the symbolic link.
Steve
Scott,
Thanks for the info. I have a solution to avoid the use of the link, but Infor has indicated that it may be a bug in 6.1.1. One of my developers identified the package require for clsqlite (replacement for gdbm) in $HCIROOT/tcl/libl/cloverleaf/init.tcl. It references the sqlite3 package that you mentioned. Everything works fine if I comment out that line. Obviously that isn’t something that I want to do. It would be nice to have it working as it was previously until I can fully implement the long-term solution.
I’ll post back to the forum when I have an answer from support/dev.
Thanks.
Steve
That’s not what i wanted to hear. 🙁
Has anyone else run across an issue using a symbolic link to hcitcl in CIS6.1? We’ve been using this for the past several verions without issue.
Thanks.
Steve
Bob,
Did you ever get this resolved? I’m running into the same issue and it just isn’t making any sense to me.
$ pwd
/appData/local/bin
$ ls -al hcitcl
lrwxrwxrwx 1 hci prod 37 Aug 18 09:45 hcitcl -> /quovadx/cis6.1/integrator/bin/hcitcl
$ ./hcitcl
application-specific initialization failed: can’t find package sqlite3 3.7
hcitcl>exit
$ $HCIROOT/bin/hcitcl
hcitcl> “works just fine”
$ which hcitcl
/appData/local/bin/hcitcl
Thanks.
Steve
Rob,
I encountered this issue as well. You need to be running at least AIX6.1 TL8. The top command that is provided by Cloverleaf is compiled for this version of AIX.
Thanks.
Steve
Not sure if this is the same on Redhat, but on AIX I source hci’s profile for each cronjob.
0 1 * * * (. /home/hci/.profile ; scriptToRun.tcl) >/dev/null 2>&1
Hope this helps.
Steve
Gretchen,
We started using the master site several years ago on 5.7 (AIX). It was applied to existing installation with no issues. You are correct that objects (scripts, tables, xlates, etc.) at the site level will take precedence over master site.
There’s no need to restart the server to make master site objects visible (aka appear in the GUI) to the other sites. You just need to make sure you update the rootInfo file (located in $HCIROOT) with the name of your master site:
mastersite=
masterSiteNameThe same rules apply when updating an object in the master site just as they do in an existing site. You must bounce, reload, or purge cache depending on what is being changed.
I haven’t experienced any issues with the master site after an upgrade. It gets copied just like any other site.
Hope this helps.
Steve
Rob,
Cloverleaf support indicated that they never tested with MQ clustering. I tried a variety of connection options, but was never able to successfully connect. I did ask that this be considered in a future release.
Hope you have better luck than I’ve had. Please let me know if you’re able to get this working.
Thanks.
Steve
Rob,
We have used aliases in the past, but our MQ Admins are the ones that do the setup so I can’t be much help in that department. I do know that this error can occur with a cluster queue. The problem with that is Cloverleaf does not support MQ clustering.
Sorry I couldn’t answer your specific question, but I did want to pass along the info about the clusters as it was frustrating to not be able to get it to work.
Thanks.
Steve
-
AuthorReplies