onLoad times are incorrect sometimes

See the attachment. I think this is a bug with chromium, but if they are not able to solve it, we need a way to detect and resolve this in mobster. One potential approach is to look at the start time of the page requested and ignore stuff before it (in the har/waterfall)
Interesting, do the 2 files at the top of the waterfall actually belong to the request?
Yes.
Those 2 calls are usually at the very end on this page.
Potentially an issue with chromium: http://code.google.com/p/chromium/issues/detail?id=141971
One of the potential issues could be related to how much time is mobster waiting after onLoad before starting the next iterations. If we don't wait long enough, the javascript calls after onLoad might pollute the timings in the next iteration.
I agree. It would be great if we could simply kill all javascript activity after determining that the page has "loaded" but after skimming the webkit protocol it doesn't look like there is a way to do it.
On Tue, Apr 9, 2013 at 4:46 PM, RiteshMaheshwari [email protected]:
One of the potential issues could be related to how much time is mobster waiting after onLoad before starting the next iterations. If we don't wait long enough, the javascript calls after onLoad might pollute the timings in the next iteration.
— Reply to this email directly or view it on GitHubhttps://github.com/linkedin/mobster/issues/28#issuecomment-16147456 .
Never mind, that wouldn't work either...
On Tue, Apr 9, 2013 at 10:13 PM, Arman Boehm [email protected] wrote:
I agree. It would be great if we could simply kill all javascript activity after determining that the page has "loaded" but after skimming the webkit protocol it doesn't look like there is a way to do it.
On Tue, Apr 9, 2013 at 4:46 PM, RiteshMaheshwari <[email protected]
wrote:
One of the potential issues could be related to how much time is mobster waiting after onLoad before starting the next iterations. If we don't wait long enough, the javascript calls after onLoad might pollute the timings in the next iteration.
— Reply to this email directly or view it on GitHubhttps://github.com/linkedin/mobster/issues/28#issuecomment-16147456 .
I guess my question is (I haven't looked at the code), do we wait X seconds after onLoad before starting the next iteration? If so, lets make it 2X or 3X and see if that fixes it.
Ritesh, Arman
If it helps we can get on a conf call to discuss the issue. ( might be faster )
thanks Hari
From: RiteshMaheshwari [email protected] To: linkedin/mobster [email protected] Sent: Wednesday, April 10, 2013 6:09 AM Subject: Re: [mobster] onLoad times are incorrect sometimes (#28)
I guess my question is (I haven't looked at the code), do we wait X seconds after onLoad before starting the next iteration? If so, lets make it 2X or 3X and see if that fixes it. — Reply to this email directly or view it on GitHub.
Hey Arman,
There are 2 events of the browser which can be leveraged
- onBeforeUnload
- unLoad.
We should do all the parsing/processing after the onBeforeUnload event , do a check at unload event
$(window).onBeforeunload(function() { //do all the processing });
$(window).unload(function() { //kill all JS activity and ensure the previous step is done. });
In summary for waterfall processing , we should do it after the onBeforeUnload have fired, else they wont line up and will ensure sanity.
we can have a call to discuss in detail
Sounds good I'll send you guys an email to set up the call.
On Wed, Apr 10, 2013 at 3:04 PM, anantrao [email protected] wrote:
Hey Arman,
There are 2 events of the browser which can be leveraged
- onBeforeUnload
- unLoad.
We should do all the parsing/processing after the onBeforeUnload event , do a check at unload event
$(window).onBeforeunload(function() { //do all the processing });
$(window).unload(function() { //kill all JS activity and ensure the previous step is done. });
In summary for waterfall processing , we should do it after the onBeforeUnload have fired, else they wont line up and will ensure sanity.
we can have a call to discuss in detail
— Reply to this email directly or view it on GitHubhttps://github.com/linkedin/mobster/issues/28#issuecomment-16205374 .