nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

OpenDeploy UpGrade to 5.6 Sp1& Performance Issues

dlagore

All:
We have just upgraded from OD 5.5.1 to OD 5.6 Sp1. Prior to the upgrade we ran our deployments and they completed, on average, in about 5 minutes - give or take network bandwidth. Immediately following the upgrade we then began to experience what we would consider poor performance. The same deployments began taking anywhere from 30 minutes to 80+ minutes. The only real functionality being run within the deployment config file is a comparison on "Date Diff" -- which is set to yes -- and was not changed either b4 or after the upgrade.

All said, has anyone else out their experienced the same issues? We have been advised to use iwodcmd rather then directly iwodstart, as well as turning off eventreporting, however this has only incrementally improved our performance and not offset the increase we attribute to the upgrade.

Sincerely,
-Doug

Find more posts tagged with

Comments

nipper

Check the logs.

One thing I noticed when we made the change is that there was a number of messages like:

delete ./dir/foo.shtml [reason missing-in-src] -- SKIPPED

Basically OD was walking the directory tree to what should be deleted & then ignoring it if
do_deletes was turned off.

Made me move to file list deployment.

ANdy

Solito

Hi,
as we are also planning to upgrade opendeploy 5.5.1 to od 5.6 sp1 i'd like to have some feedbacks especially on problems - like yours - you got.

thanks Smiley Happy

Migrateduser

Hi,
What type of deployment is being run?
A remote-diff, a filelist, or a ts-based deployment?

Are you deploying content from Teamsite? From an edition area? a staging area? a workarea?
Are you explicitly using the cached /iwmnt mount point? or the uncached /.iwmnt mount point? or //IWSERVER magic path?

What is the number of files being deployed?
Has the number increased when running the deployment with the newer version versus the previous version?

No answers yet.... just want to get a sense of the facts....

Thanks!
Bill

Interwoven Engineering

Migrateduser

That doesn't sound right. It would be great if you could open a case with Tech Support (and let me know the case #) so that we can attempt to recreate the behavior with your deployment config to see what might be the cause.

Question: Is this a fan-out deployment and are all the target nodes available during the deployment?

Todd Scallan
Director of Product Management
Interwoven
t: 408-530-7167
e: tscallan@interwoven.com

Migrateduser

If you are deploying out of a read-only Teasmite edition,
try using "/iwmnt/...../EDITION" to refer to the lastest edition.

don't use //IWSERVER prefix or the uncached mountpoint /.iwmnt (note the dot)
if deploying from unchanging read-only areas. you would want to use
those only if deploying from staging or workarea, which can change on you.

Interwoven Engineering

dlagore

Thanks, we will give this a try. I will update the thread this evening and tell you what we have found out.

dlagore

Ok, so we have tried all the suggestions listed within this thread - specifically, the /.iwmnt v. //default... did not buy us any perceived benefit. Both, iwodcmd and turing eventreporting helped - incrementally - but nothing to completely offset the substantial increase we have seen based on the 5.6 upgraded. As such, we are evaluating rolling-back at this time. Thanks for everyone's help...