This is in serverStats.txt - need help to analyze these statements:
WebAppRequests: 13 Latency: 1358 Max: 31802
WebAppRequests/../../../base/ABPortlet: 13 Latency: 1358 Max: 4931
WebAppRequests/../../../base/ABPortlet/containerMCLinkedModel._pageDispatcher: 4 Latency: 2397 Max: 3921
WebAppRequests/../../../base/ABPortlet/partialSaveABC: 1 Latency: 3 Max: 3
WebAppRequests/../../../base/ABPortlet/_pageDispatcher: 4 Latency: 658 Max: 1771
WebAppRequests/../../../base/ABPortlet/_bowstreet_show_current_page: 4 Latency: 1358 Max: 4931
What does Max: 31802 milisecs mean ? , and whats its relation to Latency: 1358 ?
Each line item under ABPortlet, do not seam to add up / aggregate ? need some details on this ?
Similarly below are the SOAP requests / SoapFaults
SoapRequests: 6 Latency: 536 Max: 6366
SoapRequests/<url:IP>: 6 Latency: 536 Max: 1684
SoapFaults: 2 Latency: 30020 Max: 30242
SoapFaults/<url:ip>: 2 Latency: 30020 Max: 30021
SoapFaults/<url:ip>/computeXYZPremium: 2 Latency: 30020 Max: 30021
The SOAP faults does it mean, it was timed out after 30 secs. ?
Again, here too what does Latency: 536 Max: 6366 ms => mean and how they co-relate.
What inference to make from these statements and possible actions. ?
Thanks in advance for help.
mburati 060000VQ202582 Posts
Re: ServerStats.txt - details2013-01-02T01:59:13ZThis is the accepted answer. This is the accepted answer.The first number is the number of that type of event/request since the last server stats log entry (5 mins apart by default), so there were 13 WebApp Requests in the last n (5) minutes in that entry.
The Latency number is the average amount of time (in ms) it took for that type of action (so on average it took 1.358 seconds to complete each of the 13 requests.
The Max number is the maximum time it took for that type of action (so in that case the longest WebAppRequest in that 5 min interval took 31.8 seconds to complete). Note, in older WPF releases there was a bug that didn't reset the max # at each server stats interval so it was the max since the webapp started, not the max in the last 5 min interval. In recent WEF releases, that max should be set after each server stats interval is logged.
Similarly, in that 5 min interval your app executed/complated 6 SOAP requests, average just over half second, with a max of 6.366 seconds. There were also two SOAP Faults (which can occur on web service timeout or a SOAP fault error thrown from the web service endpoint). Given that the latency on those requests was about 30 seconds, that does seem like a timeout issue on those requests.
I hope that info helps,
The postings on this site are my own and do not necessarily represent the positions, strategies, or opinions of IBM.