Our customers experience frequent freezing when trying to perform a Search (but not all the time).
The client log shows the following InterruptedException:
10.08.2018 07:23:04 SEVERE: --> [Thread-21] com.intergraph.web.ui.searchhub.SearchHub$SearchQueueThread.run java.lang.InterruptedException at java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.reportInterruptAfterWait(Unknown Source) at java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.await(Unknown Source) at java.util.concurrent.LinkedBlockingQueue.take(Unknown Source) at com.intergraph.web.ui.searchhub.SearchHub$SearchQueueThread.run(SearchHub.java:530)
Some more information:
this does not happen all the time
it happens only with one customer, we are not able to reproduce the error
the customer uses Java 8u121, we are using Java 8u161
I assume that this error is too generic, but I thought I might give it a try to ask here.
Does anyone experience the same issue? Any hints to solve this problem?
The problem is that the customer is a big public authority with central roll-out of Java Versions. So they won´t be able to update soon since many different products are involved in a java update. Have there been any errors reported with version 8u121?
The other problem is that GMSC keeps freezing frequently but without recognizable pattern. It´s actually not even just one functionality that leads to the freezing. Sometimes it´s the search, sometimes when the user wants to make sth in the legend... (and more different functionalities). The users report that it happens sometimes every 15 minutes and sometimes not at all during one day.
Some of the users sent me their client logs, but in most cases the freezing doesn´t show in the logs. The case that I described above was the only time I have seen that the freezing actually shows in the log.
I´m quite lost about where to continue searching for the cause.
In regards to the freezing, depending the the data, connection to the server, server load etc... some times a little patience is required with GMSC as it is not frozen but maybe be processing.
Extending the strategy for find a solution to your problem as posted by mandrats, I suggest to analyse the perfomance logs of your database system - from my experience, GMSC freezes are mostly caused by the web server waiting for responses from the underlying database.
However, I would only follow this approach AFTER making sure, it's not the client's (network) infrastructure - check this by mandrats' recommandation.
We figured out now that it must be a problem with using https instead of http. We had recently switched to https. Now we switched back to http and the freezing hasn´t happened since.
Since we are supposed to use https with all our products at some point, we still need to figure out why this happens. The strange thing is that it doesn´t happen when we use GMSC and we have another customer that uses https already without any issues... We are guessing it´s some kind of network problem. I know that this is probably some internal issue, but maybe someone here has an idea where we could proceed our search?
The simplest way to trouble shoot this is to 1st draw out the network topographies and then make note of where the failing machines reside on the network.
You may have multiple topographies, physical, http, https....
Try to find a pattern such as that they all connect through the same network switch or follow a specific route to the GMSC server, or use the same proxy server.
Trial and error can then be used to determine which route or device is causing the issue.
For a more detailed analysis, the capturing of packets may be necessary to see what is happening on the network when the failure happens using a tool such as Wireshark.
