Network connection seems to get stuck (without Google services) #7638
Comments
|
Thanks for the well detailed issue report. I don't have a gapps-free device, so I can't reproduce this. My preference is to let the free software folks handle the free software development, so I've added a "help wanted" tag to the issue. |
No problem if you don't have one. Just tell me where to ship you a Pixel 2 with copperheados. |
|
Thanks for the offer @Kamkata, but I already have too many phones and not enough time. I'd prefer it if folks who believe in free software took on some development instead. |
|
It happens 10 times a day, every time the LineageOS without Google Services switches from LTE to WiFi. To cure it I kill the Signal by pressing and holding the back button on the Samsung phone. After the next run it works fine and all missing messages arrive. Reboot or cold restart don't help because LTE is always first and in some seconds switched to WiFi so it generates LTE->WiFi event and your Signal is dead. |
|
I confirm this Signal behavior for a long time now. It happens to me about once a day. My current workaround: Finish the signal process with a long press on the back button and restart. After that everything works fine again. Device infoDevice: Fairphone 2 (FP2 |
|
It happens also on iPhone and new Android devices which were not modified. Of course not so often but it is what it is. |
|
@maedoredyti Can you please add a debug log from one of the Android devices running GCM? |
|
Although the attached debug log doesn't show anything out of the ordinary, I believe the problems discussed in this thread are probably caused by two separate issues. One is the failure to keep the websocket to the Signal server alive, discussed in #6644, #6732 and maybe elsewhere. A separate issue, also discussed in #6732 (see this for example) is that on some devices the connection to the Signal client doesn't survive a change of network type. |
|
Can people with more technical insight to the issues perhaps tell us what information we need to further lock down the issues. Since the issues persists on my device I can provide debug logs and information on my device/software. The thing is that I dont know when to best capture the logs. After starting the phone? After I notice the issue? After "soft rebooting" so the connection works again? |
|
Unless this is an entirely different problem from the ones mentioned in the various other tickets, referred to in this thread, I don't think there's much more to investigate for the time being. As I've mentioned, there are probably two separate issues involved in this. One of these (#6644) is more basic in nature, since it renders Signal completely unusable for many, if not all devices without GCM, and several PRs have been submitted for it. One of them (#7723, signalapp/libsignal-service-java#53) is now in the review phase, and looks like it might have a shot at being accepted. I've been testing a patch, which addresses the other one, the one involving changes of network type (between WiFi, mobile, etc.) for a week now and it seems to be working pretty well. It attacks the problem in two separate ways by recycling the connection both when a network change is detected and also when a keep-alive request is never responded to (which is based on/inspired by signalapp/libsignal-service-java#49, which has an excellent basic mode of operation, but would likely cause more freezes, at least on my device, due to the way it's implemented). Having both methods, is not absolutely necessary, but the first is more rapid, and should leave the device without a connection for only a few seconds, when it detects a network change, while the other is more universal, but would take longer to act (somewhere between 1-2 minutes). I have a feeling the PR will have a hard time getting accepted, but in any case I can't submit it, until the first one is through, as it concerns the same functionality of Signal and needs to be based on it. |
|
i have the same problem i believe, i'm also running LineageOS without Google Services and have disabled battery optimisation. to be slightly more specific then above the normal time i notice is instead of the first tick coming up immediately there is a delay of 30 seconds or so and then i get no second tick to say it's been delived. Often linked to changing network, i.e. walking away from wifi. Happy to submit bug reports but it's hard to know if i should wait untill the problem is presenting it's self, i.e. i can't send or just before and then just after? or try to hit submit as i leave a network, any suggestions and i'll drop in some logs. this problem being solved means a lot to me so please let me know how i can help. |
|
They simply don't want to repair it because they prefer te be tracked by google play services. Snowden, the main advertiser, should be happy that his location is tracked :-) |
|
was kinda confused by moxie saying:
does he not belive in free software? |
|
Naaaaa. This is not the case. We can't really imagine how it is to be part of such a project with just a few developers. But i must say. I know alot of people that wan't contribute anything, because nobody takes care of the pr's. If they would get merged faster. Alot of people would contribute. The Problem here is again the time moxie has. He is working more than 60 hours a week |
|
"does he not belive in free software?" |
|
There is no problem in registering signal with an anonymous number. Those who really need this know how to do it. I don't think it is arrogancy, more that it is the time they share with us for free. We as the community demand way too much from 4 people working on a service like this. Don't get me wrong. I run copperheados and don't have any google shit on my phone. The developers are free to do with their time what ever they want |
|
This earlier versions of Signal were bundled with Cyanogen mod some time ago. I have never used it because it wanted registration and google services, google gmail account. I was always far from crating accounts so I have never used it. The good approach to secure communication was VoIP or XMPP with ZRTP but for the ordinary folk it was too complicated. I don't remember such problems with missed calls. Seamless boot to the Signal network (or WhatsApp) is a very good idea but there should be a way to re-register to anonymous account, not tight to any serial number inside the phone, and use it without google spy services. |
|
Yes true. I don't get it why not allowing random alphanumeric generation as username. Like you click random user and it comes something like: h1167ghUy2 This would kill it |
|
Because I smell a rat.....suspect that someone is guilty of betrayal, deception, or causing a situation to go wrong. As it should be said "Let the people who believe in free software took on some development and implement such functionality". |
|
Could you please continue your discussion in the forum. I would like this issue to remain open for further actual work on it. I am as frustrated as you are that this annoying bug still exists and GApps free Signal development seems to be lowest priority right now, but you are only making it worse. |
|
Try checking the coding if your manually fixing it or get code help |
|
@dpapavas is it possible for you to point me to the branches where both of your fixes are implemented (does this exist?)? I've been using the 'fix' by theBoatman mentioned in the last comment here: But now Signal is nagging me to update, and it's not clear what the best solution out there is, but it sounds like it's what you're working on (at least this is the one being pursued). |
|
The repositories in question are here: https://github.com/dpapavas/libsignal-service-java You can either check out the The former has been submitted as a PR and might some day be merged into Signal. The latter has not been submitted yet, as it depends on the former. Let me know if you notice anything out of the ordinary. Here are the links for each branch: https://github.com/dpapavas/libsignal-service-java/tree/netchangefix https://github.com/dpapavas/libsignal-service-java/tree/6644-sleep-fix-3 |
|
Thanks very much! I've installed the netchangefix version on my phone (well, with the latest changes to master, just because...), and it seems to be working well. I'll pop back in if I have any issues. And thanks for chasing the PR for this - much appreciated. Would be great to have usable websocket support in Signal. |
|
Issue Status: 1. Open 2. Started 3. Submitted 4. Done This issue now has a funding of 1.0 ETH (441.51 USD @ $441.51/ETH) attached to it.
|
|
@dpapavas I think I got your point. I disagree that it's the Signal developers' duty to support non-GCM devices as first-class citicens (even though I would love them to do this). They have to do testing on a finite amount of devices, and apparently no no-GCM device is part of this (yet). But I agree that Signal should care if others report that Signal functions are broken on non-GCM devices, and especially if others put a huge amount of energy and time into debugging and suggesting fixes. Besides, I hope that once your fixes (or similar logic to unbreak dangling net sockets) get merged that they're kept and not thrown away with the next round of code restructuring. So my hope would be that this particular problem will not pop up again anytime soon. You're right, as long as Signal developers don't actually test on non-GCM devices, there's no guarantee that this won't happen again. But apparently there's enough non-GCM Signal users that care and report bugs in case new problems arise. |
|
I think the linked issues mentioned early on in this issue were resolved, so I'll close this. |
|
Unfortunately, this issue isn't resolved yet. I'm running Signal 4.31.6 (beta) on LineageOS 14.1 Android without GApps (phone is an HTC One M9) and I still suffer from the same bug: when switching networks (e.g. between wifi and mobile internet), often Signal doesn't detect this switch and I message receiving is delayed literally for hours. Only thing that helps is killing and restarting the app. |
|
@greyson-signal From which version on is this fixed? The error still always occurs when switching networks (e.g. between WLAN and mobile Internet). My workaround is still: Stop the signaling process with a long press on the back button and restart. After that everything will work fine again. Thanks a lot to @dpapavas for trying so hard to get the non-GCM fixes into Signal upstream. Device infoDevice: Fairphone 2 (FP2) |
|
@greyson-signal Original reporter here. Unfortunately, even with some of the PRs merged, this issue (“network connection seems to get stuck without Google services”) is definitely not resolved on my device. There are still some PRs created by @dpapavas which have not been merged yet (signalapp/libsignal-service-java#62, #8230)! Even though the situation seems (subjectively) perhaps a bit better than when I first opened this issue, the fact is that messages still fail to arrive for minutes to hours after a network change on a regular basis! Updated device infoDevice: OnePlus One If you need anything from me (e. g. a new debug log with a more recent version of Signal), please let me now. |
|
Be advised that some change, probably introduced around version 4.30, seems to have broken Signal for non-GCM devices again (completely, not just after network changes; see #8402). I'm not sure whether it will affect all devices (it would seem the most probable), or to what extent it will affect Signal's operation, but it would probably be wise to postpone updating the app for as long as possible. |
|
LineageOS 15 and 16 without gapps, messages not sent. Latest Signal. Probably #8402 but replying here because the comment above seems relevant. |
|
@dpapavas Can you perhaps clarify what you mean by “broken Signal for non-GCM devices completely”? Do you mean just with respect to this network change issue or completely non-functional? I’m currently on 4.30.8 and haven’t upgraded so far because of potential issues with more recent versions. |
|
Perhaps it will somehow vary from device to device, but as far as I can remember, in my case it basically broke the functionality introduced in my previous patch, that woke the device up, in order to send keep-alives and keep the connection open. This would probably mean that Signal would not be able to receive messages for the most part, much as it was before my patch. If you're using 4.30.8, you should have been affected though, so either you haven't noticed it, which would be unlikely, or you're not affected for some reason, or I may be wrong. The latter would be likely, as I haven't looked much into it, except to find a quick fix, but others seem to be affected too (see #8402). (Upon closer inspection of #8402, it seems like the problem might also (or just) prevent sending messages. As I said, I'm not sure as I didn't look into its repercussions too much, just for a way of getting rid of the exception.) |
|
@dpapavas Hm, thanks. As I said before, the problem in this issue has never really gone away for me even with your partial patches that were merged into Signal, so I likely just haven’t noticed the difference. If the new problems were introduced before 4.30.8, then I’m hopefully fine (I was mainly worried about not being able to send messages). |
|
@dpapavas, thanks a lot for your research, your patches and your kind endurance in pushing that further (to get the code pulled). For sure there have been a lot of uncooperative requests from "the free software folks", but you have to be able to distinguish -- and I really expect that from Open Whisper Systems. The stuff in your context is/had been cooperative. I would be very glad if Open Whisper Systems find a way to cooperate without investing too much of their very limited resources. IMHO that is possible especially in context of free software. Perhaps some Free Software Support Board, stocked with free software people, could be established in-between Open Whisper Systems and the free software community. For sure these have to be selected/chosen by Open Whisper Systems. For sure these need to be accepted by the free software community. For sure these have to represent the needs of both sides. Since the used licenses are indeed free software licences, I would hope that it is possible to get cooperative "free software people" on board. For sure this construct works only sustainable if these people are really able to influence -- but hey, Open Whisper Systems is free too choose these. On the bottom line, mainly addressed to Open Whisper Systems (just because they can't be replaced, but there are a lot of free software people from which cooperative ones can be chosen): Please cooperate. Benefit from the community. Of course, this only works if you comprehend their needs. The most important step is already done right: you use strong free software licences (no MIT, BSD etc.). That is indeed a very strong sign (to the community) that you want to cooperate. [Some technical question will follow in next comment.] |
|
@dpapavas: |
|
I was assuming
I didn't find what the difference between the two devices was, that is, whether it was a matter of So I'm not sure whether this specific problem could be fixed that way, but I don't think it would be worth the effort in any case. If someone can troubleshoot and fix a system component, such as Using a patched version is therefore the only way I see of having a somewhat functioning Signal without GCM, at least for as long as someone is forced or willing to investigate the problems and write the patches. |
|
@dpapavas, thanks for the reply.
Did you mention which devices you have used? I can't find it neither in this issue nor in the detailed descrption of libsignal-service-java!62.
A friend of mine is using the current version without issues except the network switching problem. What issues do you mean? |
I know for sure that the whole network switching problem doesn't occur on Xaiomi Mi A1 phones with LineageOS 15 and it for sure occurs on HTC One M8 and M9 with LineageOS 14.1 (all without GApps). |
My own Xiaomi Mi 4 had the problem and a Motorola Moto3 didn't have it. Both devices were running LineageOS 14.1, without Google apps.
See here and in the next couple of messages. |
|
@mejo- , @dpapavas, thanks for your reply.
Huh, same SW!? Does that mean that something not within the OS influences that? Or do you mean it depends on some HW specific parts of the OS? I looked into the changes made to I have found this interesting commit (cherry-picked for Lineage v15): I am unsure if the improvement on some devices are caused by a change in
After that Signal "is out of sync". Device B is able to send (which A receives), but their is now acknowledgement. Messages from device A to B are not delivered. After restarting Signal on device A, past messages and acknowledgement are delivered, thus everything works again. |
I'm not very familiar with the development process of LineageOS, but I was assuming that either different ports of the same version (i.e. 14.1 for Mi4 and Moto3) could have different system component versions, or perhaps I can't see how it could be due to hardware differences, but I can't rule it out either.
The problem my patch was trying to address, was that a network change left the old sockets intact and bound to the old IP address. Signal never detected anything out of the ordinary, so it kept listening to a severed connection. As such it couldn't receive messages and notifications from the server. I believe it could send though, as there was a backup path which, after failing to send on the severed main connection, temporarily opened a new one and sent through that. I'm not really sure though and much of that might depend on the device in various ways. You might try to look through the logs after such a session and look for anything out of the ordinary. |
|
@doak I’ve used Signal on a OnePlus One with both LineageOS 14.1 and 15.1, and I haven’t noticed any difference in behavior with respect to this issue. If there were any relevant changes between LineageOS 14.1 and 15.1, they don’t seem to have affected this issue. |
|
Thanks for all your input. I have to have a closer look, it's still weird.
That's sound like the same issue I am able to reproduce with the mentioned steps. |
|
@dpapavas Are there any updates with the Signal build? Does it still fully work without GCM? |
Not sure which Signal build you refer to. If it's the official build, probably not, because it never fully worked without GCM. Patching was required, which is open as a pull request (#8230) for more than a year now, happily being ignored. Others have made similar efforts (signalapp/libsignal-service-java#70) which seem to share the same fate. See the linked PRs for more information. |
|
This is too bad. Signal could have been so much.
Wed Sep 25 10:36:40 CDT 2019 Dimitris Papavasiliou <notifications@github.com>:
…>
> @dpapavas [https://github.com/dpapavas] Are there any updates with the Signal build? Does it still fully work without GCM?
>
Not sure which Signal build you refer to. If it's the official build, probably not, because it never fully worked without GCM. Patching was required, which is open as a pull request ( #8230 [#8230] ) for more than a year now, happily being ignored. Others have made similar efforts ( signalapp/libsignal-service-java#70 [signalapp/libsignal-service-java#70] ) which seem to share the same fate. See the linked PRs for more information.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub [#7638?email_source=notifications&email_token=ALKOUNXLMYKCXEVP2DUK3UTQLOAQJA5CNFSM4EY6LAGKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD7SLBSA#issuecomment-535081160] , or mute the thread [https://github.com/notifications/unsubscribe-auth/ALKOUNRJE7LD2XRMEZFNLBLQLOAQJANCNFSM4EY6LAGA] . [https://github.com/notifications/beacon/ALKOUNU6YG7U4W5EBGH7GODQLOAQJA5CNFSM4EY6LAGKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD7SLBSA.gif]
|
|
This bug still happens, with 4.50.5 with GApps / Android 6.0 / Huawei EMUI 4.0.3! Does anyone care that it makes Signal halfway unusable, or is the project dead? :/ |
|
I believe this bug was fixed in e3b66dc. @navid-zamani This thread is specific to devices without GApps. It's unclear what your specific symptoms are, but you may be experiencing different problems related to delayed notifications. Check out #8692. Also, be sure to first check out support page on fixing delayed notifications: Huawei is known for being super aggressive with battery optimizations, leading to things like delayed notifications.
Feel free to check the commit history :) I think the team is doing great work. |
|
I hope that my experiences with the Razer Phone 2, running stock Android 9, with Telegram, and my suspicions thereof, may help, even though my issue is with another ecosystem (I do not run Signal, due to my disappointment with it's claim of being open-source, while at the same time the Signal company is aggressively want a war against fully-FOSS forks, that remove reliance upon Google's Play Services (a bad choice for a company that supposedly cares about privacy (no offense to the engineers, who are doing a splendid job)). I am having the same exact issues as the OP, except on a Razer Phone 2, running stock Android 9 (temporary, this is not really a daily driver until I manage to boot, well anything, other than the stock ROM), with most of the apps disabled (especially Google's Play Services, including all of it's respective parts), and I keep having issues getting Telegram connecting, when it is in the foreground (I do not make use of it in the background), and yes, all battery optimizations are disabled, with unlimited data allowed (testing with WiFi, no SIM inside yet). Trying unofficial modded versions, as well as alternative clients, appears to help a little, but within the margin of error, so could be placebo. This only happens when attempting to connect to any "normal" wireless router, on both 2.4Ghz and 5Ghz (both N and AC on 5Ghz), however connecting to a mobile hotspot works without issues. From what I can clean, this is not due to battery saving shenanigans in the stock ROM either, so I can only chalk it up to the specific Qualcomm driver and it's configuration in the stock ROM (I may be wrong, this is just a suspicion). I also have an issue with uploads, which I believe is connected, where upload only upload ~120-240KB every 5 minutes or so when attempting to upload via most apps (including the browser). To be clear, it's not that Telegram does not work at all, through the browser is just fine, but rather it's connection on this ROM is utterly unstable, and sometimes I am forced to reboot just to get new Telegram messages to come in. |
and others
Re-filing #7420 due to #7598.
Bug description
This is a revival of #7420 and may be the same issue as #6447, #6644 and #6880. I am using the official Signal APK from https://signal.org/android/apk/, so not Noise or anything like that.
For months now, Signal on Android’s network connection seems to “freeze” sporadically (once or twice per day), leaving it unable to send or receive any messages (even things like read confirmations) for long periods of time. During this time, messages can be sent and received from the Signal desktop client normally and without any delay, but none of these messages show up in the Android client. It does not make a difference whether the Android client is open or any conversations are opened – nothing happens.
It seems that Signal can be made “un-stuck” by resetting the phone’s network connection (e. g. disabling WiFi so that the phone switches to cellular data) or using “force stop” to kill and restart the Signal Android client completely. Upon one of these events, all of the messages from the intervening time period will appear in the Android client all at once.
Steps to reproduce
Unfortunately, since this seems to happen erratically, concrete steps to reproduce are a bit difficult to determine.
Actual result: No new messages (sent or received) are displayed in the Signal Android app, even if it is open. This occurs when the phone and the desktop are on the same WiFi network, so the issue is not due to poor connectivity.
Expected result: Messages should appear immediately in the Signal Android client.
Device info
Device: OnePlus One
Android version: 7.1.2 (LineageOS 14.1). I have never had Google Play services installed and Android Doze is disabled (“Battery optimization: Not optimized”) for Signal.
Signal version: 4.17.5 (but I’ve been having this issue for months now, so includes many earlier versions)
Link to debug log
https://gist.github.com/anonymous/b1f29ef755a08091528fefff299057d9
At 15:25, my phone switched to cellular data (because I went out of WiFi range). At that point, all the messages from the last hour or so appeared all at once.
This debug log was taken with Signal version 4.15.5, but the issue still occurs with the latest version (4.17.5).
The text was updated successfully, but these errors were encountered: