[
https://issues.apache.org/jira/browse/CLEREZZA-432?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13000150#comment-13000150
]
franco fallica commented on CLEREZZA-432:
-----------------------------------------
Just to show you that this issue is important: My story of this morning.
On Friday night one of our clerezza instance went down (planed) for backup. It
came backup after the backup ended successfully, but it was never reachable
over the web because it has always thrown a "too many open files exception"
coming from the storageProvider. Even I am aware of this issue it still took me
30 min to find out that this was because of the cached webresource graphs.
In the end the result of this issue was that this particular clerezza instance
was down (meaning not usable by the customer) for about 55 hours !
> when starting clerezza platform a new web-resources-cache.graph will be
> created with timestamp in file name
> -----------------------------------------------------------------------------------------------------------
>
> Key: CLEREZZA-432
> URL: https://issues.apache.org/jira/browse/CLEREZZA-432
> Project: Clerezza
> Issue Type: Bug
> Reporter: Tsuyoshi Ito
> Assignee: Reto Bachmann-Gmür
> Priority: Critical
>
> http%3A%2F%2Fzz.localhost%2Fweb-resources-cache.graph1298384804817
> http%3A%2F%2Fzz.localhost%2Fweb-resources-cache.graph1298369686808
> In our case one web-resources-cache is 192MB (operating system: windows 7).
> Disc space is not expensive but I am wondering why this is necessary
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
