asked 2015-04-27 19:05:35 +0300
I suspect LowPowerMode and/or powersave mode is/are not working properly with 1.1.4.28. Since I upgraded, the battery drain has obviously been increased dramatically (confirmed by Hunger Meter and SysMon).
Same for everybody or just me?
update: seems to be fixed in 1.1.6.27
answered 2015-05-20 15:03:03 +0300
Thanks a lot to all which helped to narrow down this issue.
According to our investigations disabling multicast filtering on the wlan driver in 1.1.4.29, is causing in networks with a lot of multicast traffic, the battery drain. Due to incoming multicast traffic, more interrupts and events are triggered and prevents the 2nd core in the cpu almost always to go to suspend mode. So it really depends on the WLAN network (IPv4 and/or IPv6, amount of multicast traffic, etc) on how much of battery drain you observe.
We disabled multicast filtering because in IPv6 networks, router advertisements are delivered through multicast-messages. And the filtering caused losing ipv6-connectivity when the RA expires. E.g in our own WLAN there is an RA multicast message every 10seconds (seems other admins configure it to do every 2s!), which is quite normal on IPv6 enabled networks.
So current plan as of today 20. May 2015 (which might still change):
Due to the lack of IPv6 being very popular (still in 2015 unfortunately, but that's an other topic) and rating the battery drain being currently the bigger problem, we decided to enable multicast filtering again in the next update (1.1.4.29 + 1) as an interim step. For a later update (probably 1.1.4.29 + 3, but don't take that a as promise, it's my best guess with the info currently available) we plan to have a setting, so it will be possible to chose from UI, if multicast shall be filtered or not.
The phone could also automatically (de-)activate multicast filtering depending on its current IPv6/4 connectivity. So maybe a trifold choice like on/off/as-needed might be welcome, too.
spacefrogg ( 2015-05-20 15:26:15 +0300 )editWhat does "in the next update" mean? You say (1.1.4.29 + 1), so is it delivered? The batery drain is the only thing which holds me from upgrading my phone to the new version.
NuklearFart ( 2015-05-20 16:39:31 +0300 )editAs cpcompany I observe the drain also with WLAN deactivated.
Jolla-be really different ( 2015-05-21 07:08:32 +0300 )editanswered 2015-05-12 22:34:07 +0300
This post is a wiki. Anyone with karma >75 is welcome to improve it.
As there's no comment from Jolla Sailors to either confirm or deny if there's a bug, let's try DIT on this:
Please fill the linked form both if you have, or if you haven't experienced problems with power consumption. Both cases are important to report, to help hunting this possible bug down
UPDATE: Already after 17 reports it seems that if there is a bug, the only common finding is in different internet connections (so, at least not related to any apps or 3rd party services or combinations of those). Thanks for your quick and efficient response, please go on. Together we can achieve whatever we want on this boat! Let's enter STAGE 2:
As above, please fill the linked form both if you have, or if you haven't experienced problems with power consumption. Both cases are important to report, to help hunting this possible bug down
UPDATE: THE RESULTS
Assuming all the posts were honest and accurate, the reports posted on stage2 are not addressing this possible bug directly into any certain set or settings of connections, but there's a clear indicator that it's more common when there are different connectivity issues (28 vs 4), and it's slightly more common when the phone changes between different mobile networks (64% vs 32%).
Also, our active reporting on stage1 (thanks!) has clearly cut out a lot of possibilities, including any apps or 3rd party services and any combinations of those. This gives a hint that the power loss is in the system itself, but it's clearly not affecting all phones. For further study, we'd need to find a common factor in these phones - it might be for example full memory allocation space, more active use of swap
My best guess at this point is that there's something preventing the phone to enter sleep mode, and this is somehow related to connections in use. Also, large lists to sync might play a role.
Is there a bug? I'd say yes, something seems to be causing extra battery drain on several devices. However, don't take this as an answer. Looking forward for comment from Jolla to follow. Thanks for everybody!
answered 2015-05-13 00:39:44 +0300
i found some strange drain using android apps in combination with audio output via bluetooth a2dp. maybe there is a problem with aliendalvic connecting audio to pulsaudio?
also i forgot to mention, that i live in an area with lousy cell coverage. most if the time there is 2g reception with a maximum of 2 bars. the wifi which a can use is also not good. so there might also be some problem with the connectionmanager switching the networks. the connection manager was never good enough to switch the networks mir than a view times before i had the restart the networkservices. but after the update id happens more than once a day. more like 10-15 times a day, that i have to restart the networks. hope this can help.
answered 2015-05-22 13:56:30 +0300
The multicast issue could fit my observations well. As long as I stay in my (countryside) office, my phone will behave for days on end. When I visit the city I have had numerous crashes and the 65%->4% quick transitions reported elsewhere.
Sniffs of pin conflicts or such though, because a number of times a cold boot has not been enough, but only removing the battery for a while has made booting possible. Like two different issues actually: one excessive and the other catastrophic drain.
I stopped using the "Stop Android support" feature on May 13. Before that, catastrophic and quick battery drain was pretty frequent. Since then, not one case.
jollakin_on ( 2015-06-02 13:14:25 +0300 )editAsked: 2015-04-27 19:05:35 +0300
Seen: 4,152 times
Last updated: Jun 10 '15
please vote also here:
rudi ( 2015-04-27 19:25:20 +0300 )edithttps://together.jolla.com/question/89804/release-notes-114-aijanpaivanjarvi-early-access/?answer=90946#post-id-90946
seems to be related
Yes. However I wanted to create a dedicated post with relevant tags, as this could be a major UX issue, if confirmed.
objectifnul ( 2015-04-27 19:30:02 +0300 )editI'm also experiencing that
luchmhor ( 2015-04-28 11:08:55 +0300 )editI do not have evidence but I have the feeling phone battery now does drain faster, I would imagine improvements in smoothness and snappiness do not come for free?
pmelas ( 2015-04-28 11:27:27 +0300 )edit@pmelas Sure. Maybe we could at least have a choice. For example disabling booster-qt5.service when powersave mode is engaged.
objectifnul ( 2015-04-28 11:44:51 +0300 )edit