[Jonathan Christian]

Hey Charlie –

As always, thanks so much for your willingness to help.

I had a feeling it might be a permissions issue as well. So, I logged into the server as hciuser, and ran it both ways with the same result. Wouldn’t this have produced a similar error when attempting to run the proc in the testing tool?

I also checked the Windows permissions for the database in question, and hciuser has admin privileges against that file.

I’m sure it’s something really simple that I’m missing here, but I just can’t seem to figure it out. All of the other SQLite databases I’m reading/writing to and from work like gangbusters.

[Jonathan Christian]

Hey Jim!

Sorry for the huge delay in response. The flu attacked my household ferociously last week, and had me out of commission until today.

We’re actually on 6.0.3 right now, so if I can avoid TCL altogether, that’d be fine by me. Would you mind giving me some details on the configuration? I’ve looked around here on Clovertech for details, but haven’t dug up anything specific enough for me to go off of.

[Jonathan Christian]

Hey Levy –

Thanks so much for the response. You’re correct that I’m using the GUI to test my TPS. I’ve actually been testing this out with sms_ib_reply in the Caller Context, to no avail. I’ve attached a screenshot of my TPS test config, and the subsequent Windows error that pops up when I run the test for some additional detail.[/img]

[Jonathan Christian]

Any takers?  🙂

[Jonathan Christian]

Hey Jim –

I hate resurrecting old threads like this, but I’m having a similar challenge, and can’t find anything in the forums as a direct solution.

My particular case stems from a need to break up an OBX-5.5 into additional OBX segments if the character length in OBX-5.5 is greater than 200 characters in length. Each accompanying OBX segment would have a 200 character limit on its OBX-5.5 as well. For example, I’d need to break up the following segment:

OBX|11|ED|PDF^PDF||^AP^^Base64^JVBERi0xLjMKJeLjz9MNCjEgMCBvYmoKPDwKL0NyZWF0aW9uRGF0ZSAoRDoyMDE1MDcyODA5MDYwNSkKL0NyZWF0b3IgKEFjdHVhdGUpCi9Qcm9kdWNlciAoQWN0dWF0ZSBYTUwgdG8gUERGIENvbnZlcnRlciAxLjApCj4+CmVuZG9iagoyIDAgb2JqCjw8Ci9UeXBlIC9QYWdlCi9NZWRpYUJveCBbIDAgMCA2MTIgNzkyIF0KL0NvbnRlbnRzIDMgMCBSCi9SZXNvdXJjZXMgNCAwIFIKL1BhcmVudCA1IDAgUgo+PgplbmRvYmoKMyAwIG9iago8PAovRmlsdGVyIC9GbGF0ZURlY29kZQovTGVuZ3RoIDYgMCBSCi9MZW5ndGgxIDE3Mjk2IAo+PgpzdHJlYW0KeJztXOtv2zgS/+6/gsABhwRoZJF6F4sD2jx6WSSpm2T3sNjsB8VmbN3q4ZXk5LJ//c2Q1NNyJDvZbFq0RRs/yHnxN8PhaBhKfhxRosPfA/zheIxMo9EfI/zEpkzTdYofip8pH/2HxPDlx2vxva6ZnqsbHknno/EJJS65vquIWYalwVzH0myTX

Into:

OBX|11|ED|PDF^PDF||^AP^^Base64^JVBERi0xLjMKJeLjz9MNCjEgMCBvYmoKPDwKL0NyZWF0aW9uRGF0ZSAoRDoyMDE1MDcyODA5MDYwNSkKL0NyZWF0b3IgKEFjdHVhdGUpCi9Qcm9kdWNlciAoQWN0dWF0ZSBYTUwgdG8gUERGIENvbnZlcnRlciAxLjApCj4+CmVuZG9iagoyIDAgb2JqCjw8Ci9UeXB

OBX|12|ED|PDF^PDF||^AP^^Base64^1IC9QYWdlCi9NZWRpYUJveCBbIDAgMCA2MTIgNzkyIF0KL0NvbnRlbnRzIDMgMCBSCi9SZXNvdXJjZXMgNCAwIFIKL1BhcmVudCA1IDAgUgo+PgplbmRvYmoKMyAwIG9iago8PAovRmlsdGVyIC9GbGF0ZURlY29kZQovTGVuZ3RoIDYgMCBSCi9MZW5ndGgxIDE3Mj

OBX|13|ED|PDF^PDF||^AP^^Base64^k2IAo+PgpzdHJlYW0KeJztXOtv2zgS/+6/gsABhwRoZJF6F4sD2jx6WSSpm2T3sNjsB8VmbN3q4ZXk5LJ//c2Q1NNyJDvZbFq0RRs/yHnxN8PhaBhKfhxRosPfA/zheIxMo9EfI/zEpkzTdYofip8pH/2HxPDlx2vxva6ZnqsbHknno/EJJS65vquIWYalwVzH0myTX

Would you mind elaborating on how you’d accomplish this in an XLATE?

Thanks, as always, for all of your help!

[Jonathan Christian]

Oh, good grief! I swear I did the exact same thing before, and it didn’t work. Lo, and behold… I try it this time, and it works without a single problem. Haha

Thanks, Brent. I don’t know what black sorcery you just performed, but it worked. Much appreciated!

[Jonathan Christian]

Just as a way of bumping this thread to shed a bit of visibility to those perusing the forum, I’m also having this exact same issue in 5.8.

Anyone out there have any ideas as to why this may be occurring?

[Jonathan Christian]

Wow. I could just slap myself. That was so obvious.

Thanks for letting me know, Jim. I appreciate it.

[Jonathan Christian]

I do indeed. Mailto is set the fifth argument passed into the korn shellscript from the exec line. In this case, it’s simply my email address (Jon_Christian@chs.net).

Here’s the exec command with all of the arguments in square brackets this time, for readability:

Code:

$HCIROOTsbinemailAlert.sh [Facility001] [192.168.1.1] [site_01] [”%A”] [Jon_Christian@chs.net]

And here

[Jonathan Christian]

Hey Robert,

Here’s what I’ve got in the alert exec line:

Code:

$HCIROOTsbinemailAlert.sh Facility001 192.168.1.1 site_01 “%A” Jon_Christian@chs.net

And here’s where I’m calling the mail command in my shell script:

Code:

mail -s “$subject” $mailto < $fname

I’ve tried putting an ampersand after both of those lines, but I’m still getting two emails for each time the alert is triggered. Am I doing something wrong here?

[Jonathan Christian]

I believe I’m seeing the same issue, myself. When attempting to trigger a UNIX shell script from the alert exec for threads that go out of UP proto status for longer than 30 minutes, the engine triggers the same exec sequence twice. So instead of receiving ONE alert email kicked off by the shell script, we’re getting two.

Here’s what my alert looks like:

Code:

{VALUE pstat} {SOURCE {test_thread_in}} {MODE actual} {WITH -2} {COMP {!= up}} {FOR once} {WINDOW {* * * * * *}} {HOST {}} {ACTION {{exec {emailAlert.sh  [Facility] [IP_Address] [Site] “%A” dummy@email.com}}}}

This doesn’t happen when it’s setup to trigger against queue depth, but it does when I set it against proto or engine status.

I’ve spun my wheels on this for hours, and can’t seem to find the silver bullet to slay the beast. Any assistance on this would be greatly, greatly appreciated.

Thanks in advance!

[Jonathan Christian]

Having recently implemented alert messaging between Cloverleaf and BMC Patrol, I thought I’d pass along the details on how we accomplished it for anyone interested in doing the same.

1. Installed MSEND on the alerting box under $HCIROOT/temp/bmc (highlighted in orange in step 2).

2. Added a new alert (in our case, thread status) using the exec alert action, utilizing the following command:

<span style="color: orange]$HCIROOT/temp/bmc[/color]/msend -n @111.111.11.11″>#mc -a <span style="color: green]PIM_EVENT -m “Cloverleaf Alert –

The IP address and port information for the Patrol box the alerts are sent to are highlighted above in red, and message text itself are contained within quotations highlighted in blue.

3. PIM_EVENT was then setup within BMC Patrol as a new alert class specific to Cloverleaf triggered alert messages.

4. Tested and verified the alert as normal with the assistance of one of our Sys Admins.

Additionally, this was accomplished with Cloverleaf 5.4.1 on the AIX 5.3 platform. Also, you should be able to pick up msend through the BMC website.

This is kind of the streamlined version of what we did, so if anyone needs additional details about it, please feel free to ask.

[Jonathan Christian]

Hey, Dan. I think the idea of Visio documentation would be a great addition to our team. Would you mind sending me a copy as well, please?

jon.christian@ardenthealth.com

[Jonathan Christian]

Eureka! That did the trick. However, I feel that I’ve found the root culprit in the following error.

Proto Err: Fri Oct 12 12:07:35 CDT 2007

Error Msg: cannot open ……datainbound/OBGYN_CAD_db_20071004.BAK for reading: No

This error (and others similar to it) have cropped up before during testing and development of this interface in particular, but they haven’t blocked the flow of data into Xlates. Now though, the process is back up and running, but the inbound thread is still showing the preceding error.

We’re still pretty new at trying to troubleshoot these issues, so we haven’t really gotten our “sea legs” yet.

Thank you all, again, for your input and suggestions. They are greatly appreciated.

[Jonathan Christian]

Hello All,

I’ve run through all of the suggested solutions, with no success yet. We’ve even bounced the server to make sure we’ve covered all of our bases. It only seems to be occurring with this one process.

Additionally, I’ve dug a bit deeper and have found the following that might be of use as well…

Reported at: Fri Oct 12 09:26:32 CDT 2007

Status

[Author]

Replies