Hi
I hope someone can help with these problems.
Since upgrading our system from 6.5.2 to 6.7 we’re facing the following problems:
[ol]
[li]Website #1 - Added “ref[numbers]” objects
[/li]After a FULL publishing I found out that many “ref[numbers]” were added in the root folder. They are all empty.
All the items were published out correctly and the website works with or without them.
[li]Website #2 - FULL publishing gets stuck just before commit items
[/li]Any evidence neither in the server nor in the console log file but the Job log in Rhythmyx shows two items with “failure” status:
[ul]
Problem assembling output for item: 9-101-3583 with template: icrPgCancerTypeHighlight exception: Transaction (Process ID 61) was deadlocked on lock resources with another process and has been chosen as the deadlock victim. Rerun the transaction. see log for stack trace
[/ul]
[ul]
Problem when evaluating expression “$user.psoSlotTools.getSlotContents($sys.assemblyItem,‘icr_cancer_type_research_links’,$slotparams).size();” for variable “$link_size”: could not execute query
[/ul]
I don’t meet the same errors running a INCREMENTAL publishing
[/ol]
Any clues? Has anybody ever met anything similar and fixed it?
Thank you.
What patch level are you using? The first error sounds like an issue that we noticed with multi-threaded publishing and have since resolved in the latest 6.7 patches.
In the second error, it looks like you’ve got a " " character in the word “size” (s ize). If that’s actually what you’ve got on the template or in the binding, that could be causing the problem. If that’s not it, try breaking the call up into it’s constituent parts on the template to figure out which method is failing.
Hi
-
We’ve currently run the Rhythmyx Version 6.7.0 Build 200906R01 (3199). I’ve just re-run a full publish and this is the response: items 719, folders 57, ref[numbers] files 209. The number of the unsought files is randomly increasing…
-
I fixed this issue. There was a template which used an non-existant local field from the content type. This was only mentioned once in the server log file but the oddest thing is that it correctly works in the 6.5.2 release, no evidence in the log files whatsoever!
I keep on going searching for a solution for my remaining issue…
It doesn’t look like you’ve got any patches applied. I strongly recommend you contact TS to get the latest 6.7 patch. There are a lot of fixes in there including fixing multi-threaded publishing.
I’ve applied the RX-16101 patch but nothing’s happened…the issue is still there
What issue is still there, the folders with ref(number) being published? You’re the only customer reporting this phenomenon, so I’m inclined to believe it’s something implementation based.
Are you still using legacy XSL content lists or templates anywhere in your system?
Is this only occurring for 1 Edition or all editions?
Is this only occurring for 1 content list or all content lists?
Are there any errors generated in the publishing logs or the server.log while the edition in question is running?