Related
I'd like to know if anyone is having the same lag problems as I am getting from time to time..
The lag happens for no reason and comes at different times while doing different things.
I downloaded Watchtower from the Android Market to monitor what was happening. When the phone is running fast the /INIT process is showing as 0%..as soon as the lag starts to happen I check Watchtower again and the /INIT process is at anywhere from 70% to 100% CPU usage.
I dont use an app killer but I donwloaded one to try and see if it would correct the problem, it doesn't. The only way to solve the lag is to Restart the phone, then its fine for the rest of the day, sometimes two days but it always comes back.
I sold my SGS because of the lag issue and now it seems I have it with the DHD..
I did google this and I found that it maybe something to do with android itself..looks like /INIT is an android process..
If you notice your DHD lagging then download Watchtower and I'd bet you will see the /INIT process killing your phone..
------------------------------------------------------------------------------------------------------------------------------------
EDIT - A fix has been found to stop the /Init process lagging the phone.
On your phone go to..
Menu/settings/applications/development/now turn on Usb debugging..
This seems to keep the / init process at 0% CPU usage, it also works when the /INIT problem starts, turning USB debugging on will make the /INIT process drop back to 0%, very nice to see.
Ive contacted the Dev to find out if he can tell us why this method has fixed the problem.
Still need an update from Htc but at least we can use our phones again..
There doesn't seem to be any negative effects from turning this option on, the only thing to note is turning it off to connect to your pc could bring the /init process back and lag your phone. You can download Android Manager from the market for free which will allow you to transfer data from your pc using Wi-Fi, using this app you can keep the USB Debugging on.
-------------------------------------------------------------------------------------------------------------------------------------
EDIT 2 - It looks like you can connect to the pc via USB cable with the debugging turned on without problems, the only difference is that it will try and install additional drivers...because of the debugging mode..
Im using windows 7 and the ABD driver wont install...you need to download Htc Sync on your pc which downloads the ABD driver !!
We should also watch the battery as this may actually increase battery life...
-------------------------------------------------------------------------------------------------------------------------------------
EDIT 3 - Not one single person has come back to me and said that the /INIT problem has returned after this fix...lets hope it stays that way.. 10/11/10
---------------------------------------------------------------------------------------------------------------------------------------------------
EDIT 4 - A lot of people including myself have found that this fix improves their battery life quite a bit.
-----------------------------------------------------------------------------
Last EDIT before thread is locked just for reference and to help people !!!
For people that have the com.htc.bgp process that loads the CPU it stays at 80-90% all the time, and i have autosync on, if you stop auto sync the process goes out....
So here is the final fix in the link below (post 7)
http://forum.xda-developers.com/showthread.php?t=659325
Thank you to everyone in this thread for there efforts...amazing response.
Thanks to all the XDA Community...Including the Devs and Mods
Thread Closed
nednapalm said:
I'm getting the problem.
No apps installed except the flash and maps update, oh and OS Monitor.
Have turned off "Fast Boot" and removed battery to reset, ok for now. Very scared to start installing more apps, nearly unusable when it is lagging
Click to expand...
Click to collapse
I also tried the "turn off fast boot and it was ok for 2 days, it came back this morning though..It is very worrying and yes, the phone is unusable until it is Restarted..Thanks for the reply.
sanju007 said:
LAG WHAT?
I hope not, the samsung galaxy s have lag issus and its very bad and i sold de sgs
And now the DHD gets also lag problems ?
Ik hope not, then i will not buy this phone
Maybe an good task killer wil help ?
More people have lag issus ?
Can you make a video ?
Click to expand...
Click to collapse
An app killer doesn't solve the problem and Im unsure how many people are having this problem, to my understanding its not many...that's why I started this thread to try and find out.
I wouldn't worry yet...
I will make a video when it happens again...it may take a few days to come back....but it always does...a video added to this thread would be most useful...just have to wait till it rears its ugly head again.
So it seems that /init is a core process regarding the initialisation of the Android system and userspace hence the fact it cannot be killed using OSMonitor. It is always the first process that runs on Linux systems after boot and acts as a parent process which itself initialises then processes needed to get the OS going.
Here is a description of the Android Init Process from elinux,
'init'
A key component of the Android bootup sequence is the program 'init', which is a specialized program for initializing elements of the Android system. Unlike other Linux systems (embedded or otherwise), Android uses its own initialization program. (Linux desktop systems have historically used some combination of /etc/inittab and sysV init levels - e.g. /etc/rc.d/init.d with symlinks in /etc/rc.d/rc.[2345]). Some embedded Linux systems use simplified forms of these -- such as the init program included in busybox, which processes a limited form of /etc/inittab, or a direct invocation of a shell script or small program to do fixed initialization steps.
The init program processes two files, executing the commands it finds in them, called 'init.rc' and 'init.<machine_name>.rc', where <machine_name> is the name of the hardware that Android is running on. (Usually, this is a code word. The name of the HTC1 hardware for the ADP1 is 'trout', and the name of the emulator is 'goldfish'.
the 'init.rc' file is intended to provide the generic initialization instructions, while the 'init.<machine_name>.rc' file is intended to provide the machine-specific initialization instructions.
The Bug
The /init bug has presented itself on numerous handsets including the HTC Desire, Nexus One and Droid X models. It seems to be specific to Android 2.2 and reports the same symptoms on all handsets. Primarily this is due the 99% CPU usage causing system slowdown and left unchecked, high battery usage. It is not specific to HTC Sense ROMs as it also presents itself on stock releases. The only solution so far is a reboot. Hard Reboots and altering system settings seem to have no effect on the reproduction of this bug.
I can only speculate to what is causing the issue and hopefully a Dev can shed some more light.
Logically it seems as though a 'rogue' process is triggering /init to resume its primary function in regards to checking filesystems and attempting to spawn other system process. Due to some form of miscommunication between the processes in question, /init continues to perform these post boot tasks in a constant loop.
Think of it like this, if you try perform tasks straight after boot, the phone is unresponsive while the SD is mounted/checked and all other processes are fired up. The /init process is responsible for these tasks. If it were to loop, you would experience the symptoms we are having.
I found this on Wikipedia:
After /init has spawned all of the processes specified, init goes dormant, and waits for one of three events to happen:- processes it started to end or die, a power failure signal[clarification needed], or a request via /sbin/telinit to further change the runlevel.
So our trigger should hopefully be narrowed down to these three events.
Regarding what I posted before, the connection between sync and the bug has proven itself to be a false one.
I've also noticed that sometimes the bug can be present, however there is little impact on the responsiveness of the OS. It can seem as though everything is ok yet the CPU load can be extremely high, resulting in poor battery performance.
Anyway, hopefully more people can contribute to this thread so we can build up a better picture of what is going on. I will also try and grab a Logcat and ask anyone else near their PC when it happens to do the same.
There is a thread here which has users complaining of the same problems, yet no solutions are presented. Hopefully we can make a bit more noise and get some definitive answers.
Regards.
Hey guys some great feedback here, as with my link on starting this thread the /INIT is an android 2.2 process, auto sync on/off doesn't solve the problem neither does a hard reset, taking the battery out ect...
The guy in the link says his method worked for him, can't say the same for myself, I have a feeling his problem will be back...but if it did fix it for him then thats great, just cant say the same for me
http://forum.xda-developers.com/show...3&postcount=42
I have gone 5 days without the problem coming back and then Oh no its back, sometimes it happens twice a day, all the above doesn't fix this problem im afraid from what I have experience anyway.
The /INIT process is only 132k, something so small causes the phone to be basically unusable....very frustrating.
The 1.32.405.6 update also doesn't solve this problem, I updated yesterday and 10 min later I had the dreaded /INIT problem...
The so called "turn off fast boot doesn't fix the problem either, I have had the /INIT lag problem with it switched on and off...
What I find to be the most strangest thing is that if this is indeed an android process which it most likely is then why doesn't every DHD phone have this lag from time to time???.....why only certain phones??..
I haven't had the problem since yesterday (around 12 midday), when it occurs again I will be posting a video as requested...
The mystery continues.......There is NO solution as yet that fixes this problem.
Thanks to Lynxboy, your posts give some good information.
I wonder if this has any relevance to the problem from a linux based PC..
It seems that /INIT uses the same Process ID - 1 (found from using watchtower)
http://www.yolinux.com/TUTORIALS/LinuxTutorialInitProcess.html
BigMango said:
You purchased it from ebay, but did you buy it from a shop or from an individual (second hand)?
If you got it from an ebay shop they should give you a receipt. If it was from an individual, he should give you his original receipt (that's when the warranty started).
Anyway, the phone has a 2 years warranty, so you are covered for the next 2 years minus 1 or 2 weeks. And the paypal transaction is a valid payment receipt (it can be verified directly with paypal); so even without the receipt from the first owner HTC should cover you since the phone can't be 2 years old.
Get in touch with HTC directly, they should replace it. But the tricky part is that the problem must happen when they check it out, for this reason I would also make a video of the problem the next time it happens and send it to them by email so they can keep it on file with your RMA.
Click to expand...
Click to collapse
Thanks for the advice, Im going to contact the seller, he had it on upgrade and shipped it to me the following day..he seemed a nice guy so It should be ok...Now its just a matter of waiting for MR INIT to show up haha....wonder if he wears a hoodie LOL
bimmerfox said:
I'm not wrong because I see it with my own eyes. Even my DHD had the same problems as described. After the update no more /INIT at 90% and it keeps running smooth now.
Try to keep your patience before quoting a person who is willing to help others.
Click to expand...
Click to collapse
What?? Why do you think I made this thread...to help myself and others...as we do at XDA....the update doesn't fix the problem....just because yours hasn't lagged since the update doesn't mean its fixed....I've gone 5 days with no lag and it came back...like yours probably will.
Ridiculous....
EDIT- You can't post a statement that the problem as been fixed when numerous people have done the update and still experienced the /INIT problem...how can it be fixed??...and since the update only came out yesterday you have no way of knowing this for sure... its way to early to say that.. Try reading through posts in the forum and gather some information next time.... the update was for the youtube fix, we're not even sure that HTC were aware of the problem at the time.
Hi guys,
Just finished work and have caught up with the thread since I last posted and wanted to clarify a few things. I would also ask any new readers to find my previous post which covers much of the nature of the /init process.
The problem is solely software based. The hardware is NOT at fault!
The /init process is the parent process of Android/Linux, hence the PID 1 in OS Monitor. Imagine it as the ignition system in your car. You need to turn the ignition to fire up the other systems and that technically is the role /init takes.
It is not a rogue process in itself. It cannot be killed or closed via a task manager.
The latest update was not a fix nor does it resolve the issue.
The bug IS a Froyo bug and affects numerous devices not just the DHD although it is not present in all Froyo ROMs.
It is possible the memory management is to blame however I have run many Froyo based ROMs from FRF55 and not experienced it before; they all use the same memory management.
The three triggers for /init to awaken after its boot tasks are completed are specified in my previous post. These are the most likely causes for our issue and are in my mind triggered by some rogue process interactions.
A logcat will only really be useful if it captures the event taking place. A logcat taken when the /init process has started will report back specifically that.
Just because you haven't experienced the bug in 2days doesn't mean you are bug free, nor is there a bad batch of phones as this would indicate a hardware issue.
Continuing this thought, it is possible if you are on a branded operator ROM you may be bug free.
The /init bug is extremely annoying.
Regards.
Yes off topic...
Im asking the mod to still keep this a sticky but to lock the thread as there is really nothing left to discuss and its just here for people that are still having the /init problem as a reference...
EDIT - Thread being locked soon so....
Agian thanks to all xda community !!
Is this stickie needed anymore? Has this been corrected by any software update?
I've had this problem since the day I got the phone (December 2010) and got it again today. Once I see the obvious lag of everything, I just restart.
If it can help...
I have noticed this phenomenon the first time just after installing 2 things in my DHD :
- creating the HTCSense account (and keeping it synchronized one a day)
- installing the HTC plugin for Picasa Web Albums in the Album application.
Maybe coïncidence, maybe not.
I have disabled all HTC Sense stuff I can see. It always happens when performing some action, like opening a new app. It seems to occur so randomly, it happened again the day after my last post, but has been fine since then.
Problems with calendar after removing cache and program data.
I had major problems with com.htc.bgp using up to about 90% of my processor from time to time. When I searched the forums I found this thread.
I tried what is stated in the top post and went ahead and cleared the calendar data and the gmaill data. This is all fine, I don't seem to get any more warnings about com.htc.bgp reaching the 90% mark. One slight problem that occured though is that now when I'm syncing my calendar back to what it was, it either does not search or does not find all the other calendars that I have. The list is just totally empty. Strange as it is, It does find and sync with the default [email protected] Gmail's working fine and syncing properly though.
Since I rely on the calendar I really need this to be fixed as soon as possible.
I don't want to put words in your mouths, but I figured I might need to factory reset the phone. No biggie, but a time consumer, since I need to recustomize everything again (not rooted, not titanium).
Are there any ways of luring the calendar back on track. Jumpstarting it without a factory reset?
Thanks in advance.
Henrik
So apparently all it needed was a full night of syncing, because now it works again.
It seems though that I got rid of the annoying com.HTC.bgp that drained my processor.
Thanks to the thread
Sent from my Desire HD using XDA App
Have not experienced this since the 2.3.3 upgrade, problem solved!
I'm having this problem on my Defy after installing SetVsel, setting any values, after time the cpu stuck at 100. Watchdog lite didn't show nothing but 100%, not any process higher than 50 even. I turned on in the watchdog to show system processes and ther it showed that:
init
Linux process
Had most the time the highest percent used
I googled and found this topic. The
Menu/settings/applications/development/now turn on Usb debugging.
Click to expand...
Click to collapse
solution solved my problem. The proc isn't anymroe stuck at 100%.
I have a i9505 and I updated to android 4.4 and I have an issue with high CPU usage, I may have had the issue on 4.2 too but I am not sure.
Occasionally I notice high CPU usage and my battery life is quickly starts falling. I see two android.process.acore using a lot of CPU and I cannot kill them. When I look at them they are both com.android.process.userdictionary and seem to refuse any attempt to end the process. The only apps I had used since booting the phone was whatsapp , gosms and email app (only to view and not reply to any emails)
I have not loaded any new applications and I cannot see what has called the userdictionary process and why it keeps coming up and using high CPU.
A reboot of the device solves the issue, but it will recur again at some point.
Help?
OK so this is getting worse now.I have had t reboot the device several times today to get rid of these acore issues.
pls can someone help?
brussel said:
I have a i9505 and I updated to android 4.4 and I have an issue with high CPU usage, I may have had the issue on 4.2 too but I am not sure.
Occasionally I notice high CPU usage and my battery life is quickly starts falling. I see two android.process.acore using a lot of CPU and I cannot kill them. When I look at them they are both com.android.process.userdictionary and seem to refuse any attempt to end the process. The only apps I had used since booting the phone was whatsapp , gosms and email app (only to view and not reply to any emails)
I have not loaded any new applications and I cannot see what has called the userdictionary process and why it keeps coming up and using high CPU.
A reboot of the device solves the issue, but it will recur again at some point.
Help?
Click to expand...
Click to collapse
I'm having this problem with my NoteII LTE (GT-N7105)
Trying to find a solution - but no luck thus far.
Did you solve the issue? krtccd is always on top with up to 40% CPU usage on my Samsung Galaxy S4
vak said:
Did you solve the issue? krtccd is always on top with up to 40% CPU usage on my Samsung Galaxy S4
Click to expand...
Click to collapse
I have the same problem and I don't know the exact reason why this happens, but it is due to a mess in the swap activity of the phone. If you have a rooted phone with a kernel that let you enable or disable zram (like googy max) so try to disable it.
I think that some memory leak or some app put the memory in a bad state screwing up the swap process.
The problem should be temporary stopped if, during the high cpu activity of the krtccd process, you make a "close all" from the multi tasking manager of your phone (accessible holding the home button).
I know it is a late reply, but this could be useful if someone search the forum for a solution.
mannybiker, many thanks for your reply!
mannybiker said:
I have the same problem and I don't know the exact reason why this happens, but it is due to a mess in the swap activity of the phone. If you have a rooted phone with a kernel that let you enable or disable zram (like googy max) so try to disable it.
I think that some memory leak or some app put the memory in a bad state screwing up the swap process.
The problem should be temporary stopped if, during the high cpu activity of the krtccd process, you make a "close all" from the multi tasking manager of your phone (accessible holding the home button).
I know it is a late reply, but this could be useful if someone search the forum for a solution.
Click to expand...
Click to collapse
could you recommend some app that detects processes with a high swapping activity?
vak said:
mannybiker, many thanks for your reply!
could you recommend some app that detects processes with a high swapping activity?
Click to expand...
Click to collapse
No vak, unfortunately I have not found any app that could show me the zram used per single processes. I kept the same rom with the same config for 9 months and never experienced any problem with krtccd, then suddenly it started to happen so I think this could be related to some update the phone has made to some of my apps. The only thing I could imagine to do is a debugging period making a titanium backup of all the apps, uninstall them all, verify that the problem is solved and start restoring them periodically one after the other seeing what happen, but for the moment I have just completely removed the zram from my kernel option. It is still a controversial feature that increase available memory but also cpu and then battery consumption.
I am not a Linux expert and I have not tested this on Android, but on Linux you can try to discover the swap usage per process with something like that:
Code:
for file in /proc/*/status ; do awk '/VmSwap|Name/{printf $2 " " $3}END{ print ""}' $file; done | sort -k 2 -n -r | less
As many of you may be aware, the Z2 has an amazing battery giving possibilities of 10 hours screen on time and over a week in standby time but this is all pointless when we get wakelocks that remain active with no way to end them.
Now ordinarily I would continue this discussion within the battery performance thread but because of how large the thread has become and it is more about the performance of the battery rather than problems within android, I decided it might be better to create a thread so people can easily access.
Now we can certainly do many things to limit applications from remaining active such as stamina mode and using the greenify application (and further enhance it if root access is available) as a form of further redundancy to prevent mischievous applications running amock.
So with all this done many people including myself still have problems with wakelocks stemming from "Android System". Now I have traced two problematic wakelocks both being "handlemediaevent" and "startingdockservice", the former I have been able to trace exactly how it occurs while the latter seems to be random.
In the case of "handlemediaevent" this problem in my research I have found occurs on many other devices and the conditions for it to occur is exactly the same so this is a fault in android which i guess will need the attention of rom devs and sony to fix (this has happened in every firmware version I've used so far including custom roms). So how does "handlemediaevent" occur? It's fairly simple, when using media controls such as play/pause skip etc on an external device such as on an inline earphone control or on bluetooth controls such as on a car stereo etc so basically, anything external to the application responsible for playing media.
As for the "startingdockservice" wakelock, I've yet to reliably find what causes this has it's infrequent and random in nature, sometimes it occurs straight away after docking the phone, other times it only occurs after repeatedly docking/undocking in fairly quick succession.
Back to "handlemediaevent" how do we fix it? This is where I have stalled, I've yet to be able to find a solution for this (and "startingdockservice" too), from the other devices that have had this issue, it seems that rom updates have been key in fixing it so until then all I can suggest is to refrain from using media keys external to the phone, so if one wants to skip a track, to do it from the walkman app or whichever application is playing media, I have tested this several times playing media from bluetooth or with earphones on and not using the external controls and the wakelock never activates/becomes stuck. And when it does occur, to restart the phone, I've not been able to find a service responsible for it in the running apps/services.
So what can we do to properly fix these wakelocks? Is there a fix? Can a fix be made? Or is it a case of going to sony and ask them/wait for new firmware?
TheOnlyIntruder said:
As many of you may be aware, the Z2 has an amazing battery giving possibilities of 10 hours screen on time and over a week in standby time but this is all pointless when we get wakelocks that remain active with no way to end them.
Now ordinarily I would continue this discussion within the battery performance thread but because of how large the thread has become and it is more about the performance of the battery rather than problems within android, I decided it might be better to create a thread so people can easily access.
Now we can certainly do many things to limit applications from remaining active such as stamina mode and using the greenify application (and further enhance it if root access is available) as a form of further redundancy to prevent mischievous applications running amock.
So with all this done many people including myself still have problems with wakelocks stemming from "Android System". Now I have traced two problematic wakelocks both being "handlemediaevent" and "startingdockservice", the former I have been able to trace exactly how it occurs while the latter seems to be random.
In the case of "handlemediaevent" this problem in my research I have found occurs on many other devices and the conditions for it to occur is exactly the same so this is a fault in android which i guess will need the attention of rom devs and sony to fix (this has happened in every firmware version I've used so far including custom roms). So how does "handlemediaevent" occur? It's fairly simple, when using media controls such as play/pause skip etc on an external device such as on an inline earphone control or on bluetooth controls such as on a car stereo etc so basically, anything external to the application responsible for playing media.
As for the "startingdockservice" wakelock, I've yet to reliably find what causes this has it's infrequent and random in nature, sometimes it occurs straight away after docking the phone, other times it only occurs after repeatedly docking/undocking in fairly quick succession.
Back to "handlemediaevent" how do we fix it? This is where I have stalled, I've yet to be able to find a solution for this (and "startingdockservice" too), from the other devices that have had this issue, it seems that rom updates have been key in fixing it so until then all I can suggest is to refrain from using media keys external to the phone, so if one wants to skip a track, to do it from the walkman app or whichever application is playing media, I have tested this several times playing media from bluetooth or with earphones on and not using the external controls and the wakelock never activates/becomes stuck. And when it does occur, to restart the phone, I've not been able to find a service responsible for it in the running apps/services.
So what can we do to properly fix these wakelocks? Is there a fix? Can a fix be made? Or is it a case of going to sony and ask them/wait for new firmware?
Click to expand...
Click to collapse
I've been looking into this lately. What app are you using? Wakelock detector? I'm thinking about getting the app "Deep sleep battery" on mine soon and see if it solves the wakelock problems.
I used BetterBatteryStats to diagnose which wakelocks were active. I'm not sure about using Deep Sleep Battery as this wouldn't fix the issue, just circumvents it (if it would work).
Also you didn't need to quote my whole post, it is a long post after all.
Wakelocks for me are due to "mediascanner". My bet is on some corrupted file on my SD card. Did you install yours as a system app or just install it?
Same problem here with wakelock "handleMediaEvent". Is there still no solution?
I am more and more convinced that this is caused by some apps that do not release the wakelock properly in some cases (e.g. when the display is off). I am having the problem on a Z3C as well as i9100 when using the play/pause controls when the screen is off.
In my case, it's because of the PodcastAddict app, what app are you using when you find the handleMediaEvent wakelock ? The stock sony walkman app ? In any case, i'm pretty much convinced that although this should be fixed upstream by android (e.g. with some timeout which releases unreleased wakelocks after some time automatically), the culprit are the apps using the feature and unreleasing the wakelock unproperly.
Please refer to my post in podcast addict's feed where i found reproducible steps that cause the problem: http://forum.xda-developers.com/showpost.php?p=57767863&postcount=134
If you can also have the problem happen consistently (and to see that, you'll need a rooted device in order to run BetterBatteryStats during testing) then i suggest you contact the developer of the app to let him know.
That's still work in progress, but so far i couldn't reproduce the issue using other standard apps (Apollo, Youtube, Walkman).
Okay, is there any solution to unrelease handlemediaevent wakelock via the command or anything else without reboot? So many results in google and no one permanent solution.
TheOnlyIntruder said:
As for the "startingdockservice" wakelock,
Click to expand...
Click to collapse
I have found solution on 4PDA.ru forum.
You need to install MyAndroidTools, then go Applications -> find Settings (com.android.settings) -> click All information -> find Sevice - DockService, it should be disabled, then go Receivers and find - DockEventReceiver, and disable it.
I think I've found an apps that forces the phone to reboot, some days ago I tested a game called Clash of Puppets, at the beginning the game starts ans plays ok, but once you reach second stage it always leads to a phone reboot, maybe this can help developers to find the culprit of random reboots that CM11 users are experiencing.
Has anyone found other apps whith this vehaviour?, if yes, just post the name of your app here, maybe it will be useful.
Hope this helps quarx and others to make CM11 more stable, just my 2 cents.
Also I had a reboot because of Whatsup. Recevied a second notification while messing in settings and hard reboot.
Strange. A friend now keeps sending messages whenever he thinks I need the phone just to reboot it. No luck yet
korggis said:
Also I had a reboot because of Whatsup. Recevied a second notification while messing in settings and hard reboot.
Strange. A friend now keeps sending messages whenever he thinks I need the phone just to reboot it. No luck yet
Click to expand...
Click to collapse
Well, that behaviour is not so strange , sometimes when the phone is under low memory situation the reboot is likely to happen, the thing is... can you 100% reproduce the reboot if you do the same thing in the same app at any time? if thats the case, it would be nice to know the name of the app and how to reproduce the reboot.
Thanks for commenting, and sharing your experiences :highfive:.
sidez said:
Well, that behaviour is not so strange , sometimes when the phone is under low memory situation the reboot is likely to happen, the thing is... can you 100% reproduce the reboot if you do the same thing in the same app at any time? if thats the case, it would be nice to know the name of the app and how to reproduce the reboot.
Thanks for commenting, and sharing your experiences :highfive:.
Click to expand...
Click to collapse
I am kind of new here. But what you are saying there is just stupid in my opinion. I know, the default action when out of memory is freeze or reboot. But we are in 2014 and we have memory managers that can free up memory and take care of these situations.
korggis said:
I am kind of new here. But what you are saying there is just stupid in my opinion. I know, the default action when out of memory is freeze or reboot. But we are in 2014 and we have memory managers that can free up memory and take care of these situations.
Click to expand...
Click to collapse
would be nice if it would be that easy, but you forgot the rom as the base of any app (like any third party memory manager) is still in nightly situation. which does its own management, but sometimes it doesn't work quite right. for example because of kernel development stuff that doesn't work as it should. even a memory manager can be kicked out of background/memory from system if some low level settings are not working properly. so it is good to reproduce the issue to find out if some single apps are forcing the phone to reboot because of raising the memory usage too high too fast. even this is contribution in development. look in the development section to find out that it is simply not that easy and to find out that blaming people for trying to help is something noone needs in here! cheers
Hi, sorry, It was not my intention to blame someone I know it is not easy, just that this kind of issues should not happen too often. But you are right.
As NeoPreacher said, the phone system software is just software and can have its own flaws and bugs, in a perfect world memory managers, device drivers, schedulers and other operative systems parts would perfom always as desired but that's not the case and sometimes they can drive a device to malfunction.
I have an I9505 running lollipop and struggling to address intermittent lag.
Broadly speaking the phone starts off working well, snappy.. but after a while can start to lag. By this I mean a press on the home button may result in a 20s+ delay for the task manager to come up. Switching a task may be slow, the keyboard may take 10-20s to appear, the screen may be mostly blank for a while. It doesn't crash, and does "come back" but it's nearly unusable. Usually once switched to a task and settled down scrolling is ok. then it may work just fine, snappy for a while
I don't see oddly high figures in the various stats I can see - cpu time isn't excessive. memory is harder to interpret and at times I do see kswapd busy just after a wakeup.. so my suspicion is it's a memory leak but one I can't prove
I've tried alternative kernels - hulk, imperium, and they don't appear to help matters at all.
The lollipop build was clean & Current firmware is I9505XXUHOD7, BTU
I am rooted and have progressively disabled various packages I don't use in an effort to lighten the environment
I've disabled animations (a cause of memory leaks in 5.0.1, though supposedly not affecting samsung touchwiz)
I am a software engineer and very familar with better battery stats, gsam, even a quick peruse through dumps via adb yet still haven't got to the bottom of my issue. I'm also very aware there's little in the way of solid facts to help debug a problem in this posting......
The next step is going to be a strip back - clean install, slow addition of apps. In the meantime can anyone identify any issues they've seen with any particular apps?
planetf1 said:
I have an I9505 running lollipop and struggling to address intermittent lag.
Broadly speaking the phone starts off working well, snappy.. but after a while can start to lag. By this I mean a press on the home button may result in a 20s+ delay for the task manager to come up. Switching a task may be slow, the keyboard may take 10-20s to appear, the screen may be mostly blank for a while. It doesn't crash, and does "come back" but it's nearly unusable. Usually once switched to a task and settled down scrolling is ok. then it may work just fine, snappy for a while
I don't see oddly high figures in the various stats I can see - cpu time isn't excessive. memory is harder to interpret and at times I do see kswapd busy just after a wakeup.. so my suspicion is it's a memory leak but one I can't prove
I've tried alternative kernels - hulk, imperium, and they don't appear to help matters at all.
The lollipop build was clean & Current firmware is I9505XXUHOD7, BTU
I am rooted and have progressively disabled various packages I don't use in an effort to lighten the environment
I've disabled animations (a cause of memory leaks in 5.0.1, though supposedly not affecting samsung touchwiz)
I am a software engineer and very familar with better battery stats, gsam, even a quick peruse through dumps via adb yet still haven't got to the bottom of my issue. I'm also very aware there's little in the way of solid facts to help debug a problem in this posting......
The next step is going to be a strip back - clean install, slow addition of apps. In the meantime can anyone identify any issues they've seen with any particular apps?
Click to expand...
Click to collapse
I should add that I did previously use CM12 and that left a little laggy too, though not tried CM12.1 . The two factors holding me back - a) better/reliable camera in TW b) Beep'n'go - love the app. works everywhere unlike every other loyalty card app I've tried (uses proximity sensor pulsing, Samsung TW specific)
planetf1 said:
I should add that I did previously use CM12 and that left a little laggy too, though not tried CM12.1 . The two factors holding me back - a) better/reliable camera in TW b) Beep'n'go - love the app. works everywhere unlike every other loyalty card app I've tried (uses proximity sensor pulsing, Samsung TW specific)
Click to expand...
Click to collapse
Hi,
try to disable swap devices, if there is any. If I know well, stock Lollipop ROM uses compressed swap devices in memory, like in stock KitKat ROM (zram).
In a terminal/shell:
- List currently used swap devices:
Code:
cat /proc/swaps
- Switch to root:
Code:
su
- Disable swap devices:
Code:
swapoff /dev/swapdevice
or if you have Busybox installed:
Code:
busybox swapoff /dev/swapdevice
Where /dev/swapdevice is the swap device/partition, which you want to disable.
It's the best to do this in some kind of init.d scripts before the system starts.
I am on still KitKat+rooted, but this should work on Lollipop, too.
Good luck!
planetf1 said:
The next step is going to be a strip back - clean install, slow addition of apps. In the meantime can anyone identify any issues they've seen with any particular apps?
Click to expand...
Click to collapse
I think this is the best thing to do. I suffered a weird wifi issue that no else had. A fresh/clean install of the same lollipop ROM and kernel fixed it. Haven't noticed any apps causing lag on lollipop.