Images forgetting their content?
Adrian Williamson
Hi,
We occasionally get waves* (once or twice a year) of these errors in a stable running system:
Line 8190: [Mon Sep 23 14:43:14 2024] E_WT0029 Error at line 4313 of M2.
Line 8193: [Mon Sep 23 14:43:14 2024] E_WT0044 Object search is not defined for the current application.
So these are working OR images (11.2) often having been run for several hours that day (six hours as in the case above) where the log files show successful use of the component now described as missing.
What would cause that in a stable image?
Were the file handles closed to the image by the OS or similar?
This is on Citrix with underling SAN technology. As the location for the images and the log files they produce.
Just trying to firm up my understanding of why it happens – and file handles was all my imagination could come up with.
Cheers
Adrian
* We sometimes find perhaps 1 or 2 references to this in 1000’s of logfiles – it is the waves I am more worried about i.e. 50+ in one day.
Bodo Bergmann
- Network problems (i.e. shared drive access)
- File locked by some other process (e.g. Antivirus scanning process)
- File Handle closed by some other process
and the content of the variable is not containing a correct name (corrupted variable),
but I assume that "search" is a correct name in your application.
Not sure how line 4313 in the processed script of your M2 component looks like.
Bodo.
Bodo Bergmann Engineering Architect | OpenROAD Engineering Actian, a division of HCLSoftware | |
GESELLSCHAFTSANGABEN: Actian Germany GmbH | Registry Off: Halenreie 42, 22359 Hamburg | Geschäftsführung: Stephen Padgett, Marc Monahan | Handelsregister: Amtsgericht Hamburg | HRB 135991 | USt-IdNr: DE252449897 | |
--
You received this message because you are subscribed to the Google Groups "OpenROAD Users Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openroad-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openroad-users/000a01db0e57%244f2db220%24ed891660%24%40rationalcommerce.com.
Adrian Williamson
Hi Bodo,
This happens very rarely.
After this happens the application becomes unusable and has to be shut down.
This is all from a single image.
We do not use II_W4GL_CACHE_LIMIT.
4313 is just a straight call frame:
callframe Search(p_.....
The most recent event caused all 50 running sessions to start tossing E_WT0044 in lots of different locations across the application starting from a single timepoint.
There are references to many components as being missing which have in fact been used without issue to the point the error starts – typically this is the user pressing different buttons and trying different menus post first error.
I am sure this is nothing to do with OpenROAD; it is something that has happened in the environment it runs in.
Finding the culprit in this multi-vendor supported environment is going to be challenging.
All the users just blame the OR application.
They re-start the application and it just continues to work as if nothing happened, so they surmise that it crashed as it is ‘a poorly written’ (my words not their word) product.
I’ll run with file handles for now.
Cheers
Adrian
PS Now if I could trap the error, I could put up a message that says there has been a failure in the network infrastructure and the application needs to restart.
I see/think I saw OR12 has an attribute now about what the last runtime error was – but I really do not think I would advocate checking this after every frame navigation event / OR statement like we do with SQL statements.
From: openroa...@googlegroups.com <openroa...@googlegroups.com> On Behalf Of Bodo Bergmann
Sent: 26 September 2024 09:23
To: openroa...@googlegroups.com
Subject: Re: [openroad-users] Images forgetting their content?
Hi Adrian,
The E_WT0044 error can happen when trying to fetch the component (procedure/frame) from the image file (or database in case of "rundbapp" or running from the Workbench) on its first call. After that, it would get it from the internal source object cache.
Fetching errors could happen when there are problems with accessing the image file (or database) - this could have several reasons, e.g.:
- Network problems (i.e. shared drive access)
- File locked by some other process (e.g. Antivirus scanning process)
- File Handle closed by some other process
There would be a refetch if the component has been removed from the source object cache - see the description of the II_W4GL_CACHE_LIMIT environment variable in https://docs.actian.com/openroad/12.0/index.html#page/WorkbenchUser/Environment_Variables_for_All_Platforms.htm . Do you have this environment variable set?
The error can also happen when calling/invoking a procedure/frame/method using a variable, e.g.:
callproc :procname();
and the content of the variable is not containing a correct name (corrupted variable),
but I assume that "search" is a correct name in your application.
Not sure how line 4313 in the processed script of your M2 component looks like.
Cheers,
Bodo.
Bodo Bergmann Engineering Architect | OpenROAD Engineering Actian, a division of HCLSoftware | |
|
GESELLSCHAFTSANGABEN: Actian Germany GmbH | Registry Off: Halenreie 42, 22359 Hamburg | Geschäftsführung: Stephen Padgett, Marc Monahan | Handelsregister: Amtsgericht Hamburg | HRB 135991 | USt-IdNr: DE252449897 |
On Tue, Sep 24, 2024 at 9:56 AM Adrian Williamson <adrian.w...@rationalcommerce.com> wrote:
Hi,
We occasionally get waves* (once or twice a year) of these errors in a stable running system:
Line 8190: [Mon Sep 23 14:43:14 2024] E_WT0029 Error at line 4313 of M2.
Line 8193: [Mon Sep 23 14:43:14 2024] E_WT0044 Object search is not defined for the current application.
So these are working OR images (11.2) often having been run for several hours that day (six hours as in the case above) where the log files show successful use of the component now described as missing.
What would cause that in a stable image?
Were the file handles closed to the image by the OS or similar?
This is on Citrix with underling SAN technology. As the location for the images and the log files they produce.
Just trying to firm up my understanding of why it happens – and file handles was all my imagination could come up with.
Cheers
Adrian
* We sometimes find perhaps 1 or 2 references to this in 1000’s of logfiles – it is the waves I am more worried about i.e. 50+ in one day.
--
You received this message because you are subscribed to the Google Groups "OpenROAD Users Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openroad-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openroad-users/000a01db0e57%244f2db220%24ed891660%24%40rationalcommerce.com.
--
You received this message because you are subscribed to the Google Groups "OpenROAD Users Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openroad-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openroad-users/CAObuDgs%2Bhb0zrT4hdd2fXmS%2B0se%3DvTunb0f4jA8MevNX%3DCfXWQ%40mail.gmail.com.
Paul White
To view this discussion on the web visit https://groups.google.com/d/msgid/openroad-users/002b01db0ffd%2407ef08d0%2417cd1a70%24%40rationalcommerce.com.
Adrian Williamson
Hi Paul,
It’s a full Citrix farm running of a storage area network.
So no file shares, the Citrix server instances have a C: drive, but in reality it’s smoke and mirrors and their disk images live on the SAN.
Technology it’s great.
Cheers
Adrian
To view this discussion on the web visit https://groups.google.com/d/msgid/openroad-users/CAPzMTfA%2B%2BTn_tCi5B4x5dRJTs8uKJrRAaoMp96shFNfPeRHDFw%40mail.gmail.com.
