View Issue Details
|ID||Project||Category||View Status||Date Submitted||Last Update|
|0000540||SUMo||New Feature||public||2008-04-03 05:29||2008-05-28 19:56|
|Reporter||Zer0 Voltage||Assigned To||Kyle_Katarn|
|Target Version||Fixed in Version||1.9|
|Summary||0000540: Attempt to Reconnect on Network Failures During a Check and Support Resuming from the Point of any Check Failure|
|Description||Right now (as of version 220.127.116.11), if SUMo cannot connect to the version info server(s) while performing a check, we get a pop-up saying "Server Error. Please retry later."|
Unfortunately, there are two big problems with how this is currently working: (1) It happens too quickly and without any retry attempt and (2) You must completely restart any check from the beginning afterwards.
With regards to the first problem, a stop actually happens immediately upon any connection failure. This means that even a slight momentary network problem will cause a SUMo check to completely abort. It would be far better if rather than immediately aborting it just [automatically] retried the connection attempt again for some set period of time or number of intervals. In fact, it would be even more useful if users had a way to set what that time interval or number of retries is (although there should be some logical preset default - not 0).
The reason this failure to retry connecting is so significant is that even heavy local network utilization can cause SUMo to fail. I tend to be doing so much that this happens almost 50% of the time. I am usually at least watching streaming video, but also often have some P2P session running. In such cases my router is usually overloaded, so SUMo thinks the connection isn't available when it really is - and SUMo would keep working if it just tried to reconnect for a while.
With regards to the second problem, having to completely restart from the beginning after a check failure is highly annoying if - like with me - there are hundreds of programs in the list. If SUMo aborts at 900-something out of 1000, I'd rather not have to start over again at 1. Instead, just have it pick up where it left off later once the network connection has been restored. This would be especially beneficial in cases where there truly is some [temporary] server connection issue.
Naturally, once you release the planned version with support for saving results, it should also save the point at which a check experiences any connection failure stop (or is manually stopped). That way users could also close SUMo and then start it again anytime later and resume a check from where it stopped.
Once this support is available, I would also suggest that in cases where a check has failed or been stopped the "Check" button should temporarily change to say "Resume" instead - and then later change back to "Check" once a successful check fully completes.
Feel free to break this into two separate feature requests if necessary. I wasn't sure if I should since these issues are directly related, but I also suppose you may need to implement separate updates for each issue at different times.
|Tags||No tags attached.|
Very good idea and rich analysis.
This needs some studies on my side but this will certainly improve quality of service.
||Thanks, I try to be as detailed as possible when I report or request something. :)|
||... and I appreciate it !|
||Automatic retry (one) after 1 sec on any error|
Thanks very much for this update! Very helpful.
Any chance you would be able to make this a user controllable option in some future version?
I think it might be even more useful if there were settings options for both the number of retry attempts and the time-out value. Then users with network load issues can increase the values if necessary.
Perhaps you could add an "advanced settings" panel for settings like this - maybe for post-version 2.x low priority consideration?
||Oh, and there is still the "Resume" feature request in this. The retry update doesn't actually cover that. It would be nice to have SUMo save the check results in such a way that if there is a real connection failure SUMo can just pick up the check process later where it previously left off (rather than having to completely restart from the beginning). Thanks!|
Ok, i'll mark this one as "RESOLVED".
Please create a new demand for "Resume" which is much complex (but worth being implemented !)
|2008-04-03 05:29||Zer0 Voltage||New Issue|
|2008-04-03 20:16||Kyle_Katarn||Note Added: 0000265|
|2008-04-03 20:16||Kyle_Katarn||Assigned To||=> Kyle_Katarn|
|2008-04-03 20:16||Kyle_Katarn||Status||new => acknowledged|
|2008-04-05 01:28||Zer0 Voltage||Note Added: 0000277|
|2008-04-05 08:48||Kyle_Katarn||Note Added: 0000285|
|2008-05-26 23:43||Kyle_Katarn||Status||acknowledged => resolved|
|2008-05-26 23:43||Kyle_Katarn||Fixed in Version||=> 2.8.1|
|2008-05-26 23:43||Kyle_Katarn||Resolution||open => fixed|
|2008-05-26 23:43||Kyle_Katarn||Note Added: 0000331|
|2008-05-28 06:40||Zer0 Voltage||Status||resolved => feedback|
|2008-05-28 06:40||Zer0 Voltage||Resolution||fixed => reopened|
|2008-05-28 06:40||Zer0 Voltage||Note Added: 0000333|
|2008-05-28 06:46||Zer0 Voltage||Note Added: 0000334|
|2008-05-28 19:56||Kyle_Katarn||Note Added: 0000337|
|2008-05-28 19:56||Kyle_Katarn||Status||feedback => resolved|
|2008-05-28 19:56||Kyle_Katarn||Resolution||reopened => fixed|