DataStax Support Forums » Topic: Opscenter: Data collection appears to be working, but website itself is broken

DataStax Support Forums » Topic: Opscenter: Data collection appears to be working, but website itself is broken http://www.datastax.com/support-forums/topic/opscenter-data-collection-appears-to-be-working-but-website-itself-is-broken Software, Support, and Training for Apache Cassandra en-US Sat, 25 May 2013 22:50:59 +0000 http://bbpress.org/?v=1.0.3 <![CDATA[Search]]> q http://www.datastax.com/support-forums/search.php Anonymous on "Opscenter: Data collection appears to be working, but website itself is broken" http://www.datastax.com/support-forums/topic/opscenter-data-collection-appears-to-be-working-but-website-itself-is-broken#post-952 Fri, 30 Dec 2011 16:27:02 +0000 Anonymous 952@http://www.datastax.com/support-forums/ Got it working. In my agent config I was using hostname instead of the ip address for the local address. Thanks for your help. Anonymous on "Opscenter: Data collection appears to be working, but website itself is broken" http://www.datastax.com/support-forums/topic/opscenter-data-collection-appears-to-be-working-but-website-itself-is-broken#post-951 Fri, 30 Dec 2011 15:59:46 +0000 Anonymous 951@http://www.datastax.com/support-forums/ Turned up debug level on agent, and don’t see any errors: DEBUG [pool-4-thread-3] 2011-12-30 10:57:27,963 New CF list: ["DanCOAL" "ProdOverrideVerCounter"]["DanCOAL" "AppConfigVersionRowKey"]["DanCOAL" "MachineIDContextMap"]["DanCOAL" "DenormalizedCF"]["DanCOAL" "ProdCF"]["DanCOAL" "TagsCF"]["DanCOAL" "BranchVer"]["DanCOAL" "VerCounter"] DEBUG [pool-5-thread-1] 2011-12-30 10:57:28,329 Running real-time collection DEBUG [pool-5-thread-1] 2011-12-30 10:57:28,331 Send requested, but nothing new to send DEBUG [pool-3-thread-4] 2011-12-30 10:57:32,864 Starting process: cat /proc/loadavg DEBUG [pool-3-thread-4] 2011-12-30 10:57:32,875 Starting process: free -m DEBUG [pool-3-thread-4] 2011-12-30 10:57:32,886 Starting process: df --print-type DEBUG [pool-3-thread-2] 2011-12-30 10:57:32,928 Starting process: iostat -c DEBUG [pool-3-thread-3] 2011-12-30 10:57:32,977 Starting process: cat /proc/net/dev DEBUG [pool-3-thread-1] 2011-12-30 10:57:33,023 Starting process: iostat -x DEBUG [pool-4-thread-2] 2011-12-30 10:57:33,119 Getting metrics for DanCOAL.ProdOverrideVerCounter DEBUG [pool-4-thread-2] 2011-12-30 10:57:33,125 Getting metrics for DanCOAL.AppConfigVersionRowKey DEBUG [pool-4-thread-2] 2011-12-30 10:57:33,130 Getting metrics for DanCOAL.MachineIDContextMap DEBUG [pool-4-thread-2] 2011-12-30 10:57:33,135 Getting metrics for DanCOAL.DenormalizedCF DEBUG [pool-4-thread-2] 2011-12-30 10:57:33,140 Getting metrics for DanCOAL.ProdCF DEBUG [pool-4-thread-2] 2011-12-30 10:57:33,145 Getting metrics for DanCOAL.TagsCF DEBUG [pool-4-thread-2] 2011-12-30 10:57:33,151 Getting metrics for DanCOAL.BranchVer DEBUG [pool-4-thread-2] 2011-12-30 10:57:33,156 Getting metrics for DanCOAL.VerCounter DEBUG [pool-5-thread-2] 2011-12-30 10:57:33,255 Running short-time collection DEBUG [pool-5-thread-2] 2011-12-30 10:57:33,256 Fetching all tasks DEBUG [pool-5-thread-2] 2011-12-30 10:57:33,256 Fetching current node status/mode DEBUG [pool-5-thread-2] 2011-12-30 10:57:33,257 Fetching current drain progress DEBUG [pool-5-thread-2] 2011-12-30 10:57:33,258 Attempting to fetch old-style compaction info DEBUG [pool-5-thread-2] 2011-12-30 10:57:33,259 Getting new-style compaction status. DEBUG [pool-5-thread-2] 2011-12-30 10:57:33,260 Got new-style compaction info DEBUG [pool-5-thread-2] 2011-12-30 10:57:33,260 Fetching all outgoing stream data DEBUG [pool-5-thread-2] 2011-12-30 10:57:33,261 Send requested, but nothing new to send DEBUG [pool-5-thread-1] 2011-12-30 10:57:33,347 Running real-time collection DEBUG [pool-5-thread-1] 2011-12-30 10:57:33,349 Send requested, but nothing new to send DEBUG [pool-5-thread-1] 2011-12-30 10:57:38,365 Running real-time collection DEBUG [pool-5-thread-1] 2011-12-30 10:57:38,367 Send requested, but nothing new to send Anonymous on "Opscenter: Data collection appears to be working, but website itself is broken" http://www.datastax.com/support-forums/topic/opscenter-data-collection-appears-to-be-working-but-website-itself-is-broken#post-950 Fri, 30 Dec 2011 15:41:52 +0000 Anonymous 950@http://www.datastax.com/support-forums/ yes, it works with firefox now, thanks! however, it seems weird - I can see: All my keyspaces, etc. the data contained in my column families (and can browse them/search, etc.) which hosts are in the cluster (and it says all are online) but I can't see any of the performance stats, storage capacity, etc It also brings up a popup asking me to install agents. (which are already installed, otherwise how would I get any of this data?) In my opscenterd.conf, I didn’t tell it to ignore my Keyspaces or anything like that. In my agent logs, I don’t see any errors, just things like this: INFO [StompConnection receiver] 2011-12-30 10:31:54,212 Got new config from OpsCenter: {:http_server_port 61621, :jmx_port 7199, :metrics_enabled 1, :metrics_ignored_column_families "", :metrics_ignored_keyspaces "system, OpsCenter"} mbulman on "Opscenter: Data collection appears to be working, but website itself is broken" http://www.datastax.com/support-forums/topic/opscenter-data-collection-appears-to-be-working-but-website-itself-is-broken#post-948 Fri, 30 Dec 2011 14:58:15 +0000 mbulman 948@http://www.datastax.com/support-forums/ Ah. OpsCenter isn't currently supported on Internet Explorer. Can you try using a different browser, such as Chrome or Firefox? I'll look into the blank.html issue, but I don't think that's causing any problems. Anonymous on "Opscenter: Data collection appears to be working, but website itself is broken" http://www.datastax.com/support-forums/topic/opscenter-data-collection-appears-to-be-working-but-website-itself-is-broken#post-947 Fri, 30 Dec 2011 14:03:31 +0000 Anonymous 947@http://www.datastax.com/support-forums/ It's definitely looking for .cluster. I'm using ie8 and win 2003. When I put the URL for cluster directly I'm given a save as prompt. Opscenter is running on Rhel 6.1.2. When I load Opscenter I see host:8888/opscenter/index.html. What about the blank.html file? mbulman on "Opscenter: Data collection appears to be working, but website itself is broken" http://www.datastax.com/support-forums/topic/opscenter-data-collection-appears-to-be-working-but-website-itself-is-broken#post-946 Fri, 30 Dec 2011 13:31:33 +0000 mbulman 946@http://www.datastax.com/support-forums/ OpsCenter should never be requesting the url "/.cluster", but instead should be requesting "/cluster". Can you run <code>curl <a href="http://<OPSCENTER-HOST>:8888/cluster" rel="nofollow">http://<OPSCENTER-HOST>:8888/cluster</a></code> and see if you get proper output? For reference, can you tell me what operating system OpsCenter is running on, as well as what OS/browser you are using? Can you also paste the full url you see in your browser's location bar when you load OpsCenter? Anonymous on "Opscenter: Data collection appears to be working, but website itself is broken" http://www.datastax.com/support-forums/topic/opscenter-data-collection-appears-to-be-working-but-website-itself-is-broken#post-945 Fri, 30 Dec 2011 12:08:11 +0000 Anonymous 945@http://www.datastax.com/support-forums/ Hi, I'm running cassandra 1.0.6 with a cluster of 4 nodes. I used the tar version of opscenter free (1.3.1), untarred it on my rhel6 box. I changed the .conf file to refer to the cluster seed hosts, etc. I ran the create-agent-install and installed the agent on the 4 nodes of the cluster. I ran the setup on each of the nodes and ran the agents. Based on the opscenterd.log, opscenter is getting data from each of the nodes. (I see things like 'Processing node detail from agent for <hostx>' Now to my problem. When I go to the URL for opscenter in my browser, it comes up with a black screen that says loading opscenter, and then has what looks like 2 black frames. On the center frame, it says 'Error Loading Opscenter' and 'Error: Unable to load .cluster error code 404'). My intuition tells me that opscenter should be writing this file and isn't and therefore the webserver part of opscenter can't read it (404). The other weird thing is that there is at least one other file that in the http.log reports as 404, though it can find others. So it can't find: /opscenter/.cluster /opscenter/js/dojotoolkit/dojo/resources/blank.html (I would think this is a static file and would come with the opscenter tar?) Any idea what's wrong? Note that if I vi/touch a .cluster file and put it under the 'content' folder, the mesasge on the website changes so I know that I have appropriate permissions, etc. Note - due to employer policy I can't post logs directly. Thanks in advance.