[David Burks]

I think I have spoken to both people experiencing this, one by phone and other on a webex.  If anyone else encounters and creates an incident please mention my name in first event note.  I want to stay on top of this.

• David Burks

[David Burks]

Correct user is TEST.  Note it is all upper case.

[David Burks]

You cannot open cases via email any more.

You need to log in to inforxtreme.com and create your case there.  Once created the system will assign it to an analyst.

[David Burks]

The update referred to is 5.8.4.0.  Until that is released anyone experiencing this problem will have to use the workaround of copying the jre directory.

[David Burks]

I am attempting to duplicate this.  I have not been able to duplicate the case change.  It stays lowercase in all my testing.  

Reading through the posts it seems everyone having a problem is running Cloverleaf 5.7 on AIX.  Is that accurate or is someone also having difficulty on different OS?  Is everyone using the same version of AIX?

For those having this problem with log history feel free to send me any information about your setup and observations.  Be good if you could send me your siteInfo files.  Send to techsupport@healthvision.com with attention Jimmy and David.

[David Burks]

Robert, I would like to investigate this further.  Can you send me the gc_killMsgIfDestDown proc tcl file so I can review it?  Please send the file itself and not a screen shot or paste.

Send to techsupport@healthvision.com attention David

[David Burks]

I have sent you 3 emails on this issue.  Please check to make sure you we are not getting filtered by your spam filters.

Also, please use techsupport@healthvision.com instead of support@healthvision.com.  Using the latter eventually gets to Cloverleaf support but only after waiting it’s turn in our internal support group and recognized as something that needs to be forwarded to us.

If you still have not received anything on this then my may want to call support.

[David Burks]

Jennifer, this is really simple once you have done it once.  This should take only a few minutes to sort out and get you on track.  Why don’t you call in to support and ask for me?

[David Burks]

Posting to this older thread because someone saw this and thought they were having same problem.  I wanted to clarify some things.  The relevant error for the original poster is this one.

[dbi :dbi :ERR /0:       obhboc] Db_Vista database error -921: ‘SYSTEM/OS error: -921

Now the following error message “EXCEPTION_ACCESS_VIOLATION” was in the log but was simply a consequence of the 921.  There are many things that can cause an EXCEPTION_ACCESS_VIOLATION that have nothing to do with a 921.  Usually that is either bad data or permissions caused by some other software or program accessing the cloverleaf files while cloverleaf is writing to them.  

For the 921 what you do depends on your OS.  First we will assume this was on a unix box.  We look and see what the duplicate user was.

[dbi :dbi :ERR /0:       obhboc] Duplicate DB_VISTA user id: ‘testobhboc’

The following command could have been used.

lmclear -u testobhboc -mp

Alternatively if this does not work the following command could be used

hcilmclear -p processname

where processname is the name of the process that is having a problem.  

Now if you are on windows the lmclear command will not work because there is no lock manager to process the command.  So when you get the dbvista -921 on windows you have to stop all engines on all sites and all daemons on all sites and then bounce the cloverleaf service.  I usually hit task manager and go to the processes tab to verify there is nothing hci still left running.  If possible the service should not be stopped while any engines or daemons are running.  The exact name of the service varies slightly on each cloverleaf version so if in doubt you may need to call support to be certain.  

So in summary the EXCEPTION_ACCESS_VIOLATION may be very important when not preceded by the 921 but when the two are found together deal with the 921 and the EXCEPTION_ACCESS_VIOLATION should disappear since it is 99% probably caused by the real 921 problem.

Hopefully this clears this up a little and does not confuse anyone more.

[David Burks]

Sorry if I was not clear enough with my question.

Let’s say for an example and discussion you send an email to support and they reply with a solution to your problem.  Would you prefer the solution to your problem be at the top of the reply or would you prefer to scroll down to the bottom of the reply to find your solution?  

I might also ask if the solution does not work or you want additional clarification and you reply to the reply will your additional questions be placed at the top or will you scroll all the way down to the bottom and type your new questions at the bottom?

[David Burks]

I would also encourage you to lobby your management to send you to the training classes.  When I was first handed Cloverleaf responsibility I was a Cobol programmer.  I did not attend any classes for 6 months.  After attending classes though I was doing in hours what had took me days simply because I then knew the toolset available.  I highly encourage anyone new to the software to attend the training.  In my experience it really did pay for itself in man hours saved.

[David Burks]

I can confirm this has nothing to do with which OS or version you are running.  Mark is right.  The messages are in memory and that memory is owned by the process.  To get rid of them from memory stop the process, not the thread.  The rule I use is if you are going to delete messages from the recovery database stop the process first.  Follow that and you are guaranteed success every time.

[David Burks]

When you set up that alert you might consider configuring so it does not restart the process until it has been down for say 10 minutes and make sure everyone knows about it.  It can be a little frustrating if you are intentionally bringing something down for troubleshooting or other purposes and it keeps coming back up and nobody remembers why.  😉

[David Burks]

I have seen and am aware of what you are talking about.  I have at one time submitted a bug on it.  Unfortunately, I have never been able to duplicate it on demand for the benefit of development to prove I am not on something.  

Not sure about the jumping issue in 5.5.  I have not seen it but then I do most of my troubleshooting from the command line so not sure if that means it has been fixed or I just have not hit the exact circumstance for it to happen.

But I do have some good news for those considering the move to 5.5.  The suggestions for syncing the NetConfig and Netmonitor views is interesting so I thought I would mention that in 5.5 these views are in fact synced for you.  Both the NetConfig and NetworkMonitor GUIs use the same view file.

[David Burks]

If you are on Cloverleaf 5.4 or 5.4.1 and your GUI thread status still shows 1 hour off that means you did not update the java on your windows client per the instructions in the technical bulletin under the section labeled Windows, Linux, HP and Solaris OS platforms.

This is not an optional step.  For client boxes that do not have client only you do not have to setroot or setsite.  But you will have to download the tzupdater as directed in the technical bulletin and run it from your …integratorclguijavabin directory as instructed in the instructions.

This has been tested on 3 different boxes here and worked fine on all 3.

[Author]

Replies