Ask / Submit
8

[bug] 'Sailfish-eas' stuck at 100% CPU useage.

asked 2016-10-04 10:45:20 +0200

Beethoven gravatar image

updated 2018-09-14 14:08:06 +0200

I just moved to the Intex (wow), but I found it ran hot and checked out the 'top', and terminal shows that this process is stuck at 100% pretty much all the time, I have one exchange email account that is set up, but it's not a busy account, does this process handle things other than exchange emails? Or does anyone have any ideas as to why it could be running so hot?

Thanks!

edit retag flag offensive close delete

Comments

Your exchange account 'could' also handle your contacts and calender, depending on what you configured. What sync interval did you use? try syncing at half hour or hourly intervals this might help.

Andrik ( 2016-10-04 15:08:01 +0200 )edit

+1 I see similar issues with sailfish-eas, not always 100%, usually only around 30-40% but continuously so that my phone feels slightly warm. My current sync option is completely manual so I don't understand why it would be running warm.

stoft ( 2016-12-14 17:57:36 +0200 )edit
3

If you can enable developer mode on your device, please do the following:

1) edit /home/nemo/.config/eas-sailfish.conf so that it contains the following:

 [logging]
 Sailfish.eas.debug=dwc
 Sailfish.eas.warning=dwc
 Sailfish.eas.error=dwc
 Sailfish.easwbxml=dwc
 Sailfish.easnetwork=dwc
 Sailfish.easverbose=dwc

2) reboot your device

3) wait until you see the sailfish-eas process taking up the CPU time again

4) ssh into your device, and run: devel-su journalctl -b | cat >> eas.log

5) send that eas.log file to chris dot adams at jolla dot com and I will investigate or forward to the appropriate person for investigation.

Thanks!

chris.adams ( 2016-12-15 06:28:36 +0200 )edit

@chris.adams sent! looking at the log I get the feeling something is wrong with my contacts . My exchange account is set to sync to device only .

stoft ( 2016-12-15 12:40:46 +0200 )edit
1

Thanks for the log. We haven't analysed it fully, but a quick look showed a few things: 1) a huge jump in timestamps (8:33 to 9:38 = an hour and six minutes) when doing some processing; 2) a massive number of contacts (upward of 30,000) being reported; 3) some local modifications being reported for non-upsynced contacts which suggests that somehow the local-to-remote mapping is being corrupted. The last of these is almost certainly a bug in the plugin. The first is most likely a performance bug (e.g., some extremely inefficient processing algorithm when dealing with large datasets - we will investigate). The middle may simply be a reflection of how many contacts are in your exchange addressbook on the server - are you able to tell us roughly how many there are?

chris.adams ( 2016-12-16 09:17:52 +0200 )edit

1 Answer

Sort by » oldest newest most voted
0

answered 2018-09-14 12:32:28 +0200

updated 2018-09-17 09:37:19 +0200

I've submitted the log as with 2.2.1.18 this issue is still present and the battery get's drained in couple of hours.

[EDIT] The issue appears to be linked to contact sync, with that disabled the issue has not comeback

edit flag offensive delete publish link more

Comments

All of a sudden the describe issue appeared here as well. Note: I've been using exchange connection on all Sailfish X version since last October. All was fine till Saturday (2 days after 3.0.0.8 update). Btw. 3.0.0.5 was OK. Based on the history of this bug I'd assume it is not related to Sailvish OS version but rather to a particular situation on the contact list. I have disable contact sync for now and the sailfish-eas settled down. Will probably debut at some point later.

Valentin ( 2018-11-13 00:09:51 +0200 )edit
Login/Signup to Answer

Question tools

Follow
3 followers

Stats

Asked: 2016-10-04 10:45:20 +0200

Seen: 624 times

Last updated: Sep 17