AnsweredAssumed Answered

Alfresco generating thousands of exceptions a minute, can't find the cause

Question asked by lmitchell on Jan 30, 2014
Latest reply on Feb 3, 2014 by scouil
Hi All

I'm new to Alfresco and I've got an issue that's quite vexing.

I've installed 4.2c Community Edition on Ubuntu 12.04 using Postgres and Tomcat bundled with the install binary and everything seemed to be working until I hit on three snags that I can only assume are related:
- Whenever I login to server:8080/share using the admin account I get a popup box with "Internal server error" and an OK button (click OK and it goes away until the next sign in)
- Whenever any user tries to create a site from within the Share interface it returns the following error in a popup titled "Could not create site"
00305598 Wrapped Exception (with status template): 00300128 Failed to execute script 'classpath*:alfresco/templates/webscripts/org/alfresco/repository/site/': nodeRef is a mandatory parameter

- On checking the log files I've noticed that Alfresco is generating literally thousands of exceptions a minute, all something along the lines of:

08:35:45,129 ERROR [org.springframework.extensions.webscripts.AbstractRuntime] Exception from executeScript - redirecting to status template error: 00293586 Unknown method specified to remote store API: has
org.springframework.extensions.webscripts.WebScriptException: 00293586 Unknown method specified to remote store API: has
   at org.alfresco.repo.web.scripts.bean.BaseRemoteStore.execute(
   at org.alfresco.repo.web.scripts.RepositoryContainer$3.execute(
   at org.alfresco.repo.transaction.RetryingTransactionHelper.doInTransaction(
   at org.alfresco.repo.web.scripts.RepositoryContainer.transactionedExecute(
   at org.alfresco.repo.web.scripts.RepositoryContainer.transactionedExecuteAs(
   at org.alfresco.repo.web.scripts.RepositoryContainer.executeScript(
   at org.springframework.extensions.webscripts.AbstractRuntime.executeScript(
   at org.springframework.extensions.webscripts.AbstractRuntime.executeScript(
   at org.springframework.extensions.webscripts.servlet.WebScriptServlet.service(
   at javax.servlet.http.HttpServlet.service(
   at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(
   at org.apache.catalina.core.ApplicationFilterChain.doFilter(
   at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(
   at org.apache.catalina.core.ApplicationFilterChain.doFilter(
   at org.apache.catalina.core.StandardWrapperValve.invoke(
   at org.apache.catalina.core.StandardContextValve.invoke(
   at org.apache.catalina.authenticator.AuthenticatorBase.invoke(
   at org.apache.catalina.core.StandardHostValve.invoke(
   at org.apache.catalina.valves.ErrorReportValve.invoke(
   at org.apache.catalina.valves.AccessLogValve.invoke(
   at org.apache.catalina.core.StandardEngineValve.invoke(
   at org.apache.catalina.connector.CoyoteAdapter.service(
   at org.apache.coyote.http11.AbstractHttp11Processor.process(
   at org.apache.coyote.AbstractProtocol$AbstractConnectionHandler.process(
   at java.util.concurrent.ThreadPoolExecutor.runWorker(
   at java.util.concurrent.ThreadPoolExecutor$

The log file for any given 24 hour period is a good 100mb, so in just over a week it can generate an easy 1gb of logs - which is obviously not good and also makes it too big to upload, but the exceptions appear to just be the same thing repeating over and over anyway

Other than these issues everything seems to be functioning; users can upload, download and edit files and folders in the repository with no problems at all

I've tried rebuilding the SOLR indexes to no avail, I've checked the permissions on the alf_data folder and everything seems to be OK there as well. The other peculiar thing is that I did a fresh and identical install on a spare machine, did a cold backup and imported the PSQL database dump and alf_data files from the problem build into the fresh build (which was working fine to begin with) and the issue carried over as well.

To further compound the problem I didn't actually notice the issue until we'd already been using the build for a couple of weeks so there's now a few thousand documents stored in the problem build so starting again really isn't an option and doing a cold backup and restore into a new build just seems to take the problem along

Any help anybody can offer would be really, really appreciated