directory.cache.synchronise.interval in Jira 7.0.0.

[Krisse]

Hi, where is 'directory.cache.synchronise.interval' determined in JIRA 7.0.0. ?

/Kristian

[Charly [DEISER]]

Hi @Krisse

I believe you are looking for the LDAP directory synchronization interval, check this link:

https://confluence.atlassian.com/adminjiraserver071/connecting-to-an-ldap-directory-802592350.html

and look for the key: Synchronization Interval (minutes)

Hope this helps!

Regards

[Krisse]

Hi @Charly [DEISER]

I'am looking for 'directory.cache.synchronise.interval' from the database, so I can increase the time that JIRA uses for 'User Directory' synchronisation with Crowd. Right now it's 600 seconds and we need to change it to ~3600 seconds.

/Kristian

[Marcin Kempa]

Hi @Krisse,

I am not sure I understand your case well, but so far it looks like you would like to change the interval at which Jira synchronizes changes from Crowd to it's internal cache. Is that a fair assumption?

Is there any reason why you cannot change this value in the Jira's UI as described here?

Hope that helps,

Marcin Kempa

[Krisse]

Hi @Marcin Kempa,

Yes, I wan't to change that interval from 600sec to 3600 and I cannot do it via UI, because we have only Crowd accounts in use.

And: "You cannot edit this directory because you are logged in through it, please log in as a locally authenticating user to edit it."  ...is the problem.

So we have to do it from database and that's why I'm askin where 'directory.cache.synchronise.interval' is determined.

/Kristian

[Charly [DEISER]]

Hi @Krisse

In this case you have to edit this property in the "cwd_directory_attribute"

Make sure you are editing the correct "directory id" (first column).

I believe you are going to need to restart your Jira instance afters this change.

Regards

[Krisse]

Hi, @Charly [DEISER]

I will try that one! Thanks!

[Joni Nevalainen]

Hi @Charly [DEISER], thanks for the info. I'm using the same instance as Kristian is. The problem is that we have a large enough crowd database with aliasing enabled, and a single-threading crowd instance. Jira does not get a reply within 600 seconds from crowd when synchronizing, and the users+groups are not transferred.

We would like to increase the synchronization wait time at JIRAs end so crowd would have enough time to go through it's databases and respond.

[Charly [DEISER]]

Hi @Joni Nevalainen

I understand, in our case we are using LDAP and we have some properties like

• ldap.read.timeout
• ldap.search.timelimit

in our cwd_directory_attributes table. Probably for crowd you are going to have similar properties.

Try changing this properties (if you have them).

Regards