GUI white screen of death #257

littleskunk · 2016-06-30T20:24:36Z

Please add here any information you have. It will help us to find the reason.

matanbr · 2016-07-01T05:54:13Z

Started on the last version (0.8.7) , noticed that RAM usage is increased to around 1000MB during the first hour than the crash happen.
by the way cpu usage on this version is also higher than previous, around 30-50% all the time of use, didnt had it before.
Windows 8.1

littleskunk · 2016-07-01T06:46:24Z

That looks like https://github.com/Storj/storjshare-cli/issues/95

Anyone else with the same result?

mo35 · 2016-07-01T06:50:17Z

same as above , when it works (not always , usually i have to restart gui couple times to start receiving shards , most of times its just sits there and does nothing , background processes just running no cpu usage or ram increase , nat via upnp works) , but when it works ram usage is quite noticable , usualy when consumed ram reaches ~1gb , gui stops responding (black window) , all background processes dies. (using latest gui 0.8.7)

trippypippy · 2016-07-02T08:14:25Z

I am having the same issue although I have not noticed the RAM spike mentioned in other posts. I am using Server2012R2 and after a random amount of time running 0.8.8 windows just goes completely black and stops working (no shards). The only way of getting it to start working is to "Quit" the program and start it again. When it is actually working I get shards and everything seems ok with no unusual messages in the log.

Blackwolfen · 2016-07-03T07:57:12Z

Same issue, first it makes many shards and then becomes black. No real reason, ports and all looks good.

Blackwolfen · 2016-07-04T04:50:29Z

Had the same issue today again, i had the developer zone open and it said target disconected, and also in the log of the developer zone i saw could not send telemetric data and stuff.
That issue is getting really annyoing :)

Blackwolfen · 2016-07-04T04:51:42Z

And yesterday i saw in the developer console that around each 10 - 20 time file is in use error when it tried to access the db. Note i have only one Drive shared still it has somehow sync issues. Maybe they lead to a death lock?

ghost · 2016-07-04T07:38:07Z

After few hours of working GUI client was nothing but black window, transfers seemed to stop too, but top bar with menu responsible, restart from that menu worked.

trippypippy · 2016-07-04T08:58:50Z

Further to what Blackwolfen said I am also seeing a lot of "Failed to Collect telemetry data" with a red number to the left which I assume is the number of tries that have failed (currently 3). Eventually followed by a black screen.

littleskunk · 2016-07-04T13:07:02Z

Please open cmd / terminal to start Storjshare GUI. After GUI crash please check cmd / terminal for any error output.

If you have some debug skill you can create memory snapshots (Developer console->Profiles) . Upload them as ZIP file if you don't know how to read them.

My GUI crashed while I created memory Snapshots. I will try again to be sure.

[13624:0704/145315:FATAL:memory_win.cc(38)] Out of memory, size = 25165812
Backtrace:
        (No symbol) [0x002AC8E7]
        (No symbol) [0x002A6E27]
        GetHandleVerifier [0x00308B0E+188270]
        callnewh [0x5A84B3BE+24]
        fcloseall [0x5A82DAB8+34686]
        v8::TypeSwitch::match [0x5486BEAD+1405]
        v8::TypeSwitch::match [0x5491BF13+722403]
        v8::TypeSwitch::match [0x5491D378+727624]
        v8::TypeSwitch::match [0x54910D46+676886]
        v8::TypeSwitch::match [0x5491ACD4+717732]
        v8::TypeSwitch::match [0x5491875E+708142]
        v8::TypeSwitch::match [0x5496DE41+1058065]
        icu_54::IDNAInfo::reset [0x01DAA096+1012838]
        icu_54::IDNAInfo::reset [0x01DA9E6D+1012285]
        icu_54::RegexPattern::operator!= [0x01A43C7C+1882748]
        icu_54::RegexPattern::operator!= [0x01A4CB4B+1919307]
        icu_54::BreakIterator::isBufferClone [0x00395501+215953]
        icu_54::TimeZone::setID [0x00949C5C+5113900]
        icu_54::TimeZone::setID [0x00948FB8+5110664]
        icu_54::TimeZone::setID [0x00949E6D+5114429]
        icu_54::TimeZone::setID [0x009031B7+4824455]
        icu_54::MessagePattern::partSubstringMatches [0x02616EEE+3768542]
        icu_54::MessagePattern::partSubstringMatches [0x02588A25+3185685]
        icu_54::RuleBasedBreakIterator::operator!= [0x00BEA67D+872317]
        icu_54::RuleBasedBreakIterator::operator!= [0x00BEAA1C+873244]
        GetHandleVerifier [0x002F92D5+124725]
        IsSandboxedProcess [0x015D0FF5+163813]
        IsSandboxedProcess [0x015D081D+161805]
        IsSandboxedProcess [0x015D06DD+161485]
        IsSandboxedProcess [0x015D128F+164479]
        GetHandleVerifier [0x002F92D5+124725]
        (No symbol) [0x002C37EC]
        (No symbol) [0x002C2C84]
        GetHandleVerifier [0x002FB49A+133370]
        (No symbol) [0x002C35E0]
        (No symbol) [0x002C359D]
        icu_54::TimeZone::setID [0x0091E8FB+4936907]
        icu_54::BreakIterator::isBufferClone [0x003D288E+466718]
        icu_54::BreakIterator::isBufferClone [0x003D273D+466381]
        icu_54::BreakIterator::isBufferClone [0x003CFBC3+455251]
        (No symbol) [0x0016663D]
        icu_54::Transliterator::setID [0x0221453C+184620]
        BaseThreadInitThunk [0x743838F4+36]
        RtlUnicodeStringToInteger [0x775F5DE3+595]
        RtlUnicodeStringToInteger [0x775F5DAE+542]
        (No symbol) [0x00000000]

trippypippy · 2016-07-04T13:43:57Z

Hi The GUI crashed again but this time console was open and I got this message "Inspected Target dosconnected" Once it reloads we will attach to it automatically.

Hope this helps.

RedRockMining · 2016-07-04T14:00:59Z

Unchecked "Send Telemetry Reports". No GUI BSD for 24 hours and counting. Getting files now too, so this appears to be at least part of the issue.

byterunner13 · 2016-07-04T14:54:47Z

After few hours of working GUI client was nothing but black window, transfers seemed to stop too, but top bar with menu responsible, restart from that menu worked.

trippypippy · 2016-07-04T15:36:55Z

Hi just to confirm my post above that I get a window with "Inspected Target disconnected" "Once it reloads we will attach to it automatically." if the console window is open every time the GUI crashes.

littleskunk · 2016-07-04T19:38:07Z

I created a few RAM snapshots and it looks like to many loglines. I created a new binary with loglevel warn. If you like you can install it and try it again.
https://github.com/Storj/storjshare-gui/releases/download/v0.8.8/storjshare-gui.loglevel.warn.exe

This is not a fix. It is only a workaround and will show us if that is realy the reason for the crash. Lets see how long this version will work without a crash.

Keep in mind that you need a few restart to get a connection to a close node that will send you contracts (core issue). Start your node and wait 5 minutes. Watch for the following logline. If you don't get this line restart your node and try again.

{error} [Mon Jul 04 2016 21:36:43 GMT+0200 (Mitteleuropäische Sommerzeit)]
Contract no longer open to offers

matanbr · 2016-07-05T07:15:42Z

Well it's like you said , it doesnt fix the problem it's just delaying it.
but at least now it crashed after almost 12 hours ...

littleskunk · 2016-07-05T08:02:31Z

My GUI node with loglevel off is still running with 70MB. -> The log window is the problem. I will try to fix it.

RedRockMining · 2016-07-05T12:41:55Z

Yup, mine crashed too, overnight. Not sure how long, but before I went to bed, one instance of Storj Share had grown to around 500MB.

byterunner13 · 2016-07-05T14:56:10Z

now i'm getting the flag no tunnelers found

On Tue, Jul 5, 2016 at 8:41 AM, RedRockMining notifications@github.com
wrote:

Yup, mine crashed too, overnight. Not sure how long, but before I went to
bed, one had grown to around 500MB.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#257 (comment),
or mute the thread
https://github.com/notifications/unsubscribe/ARtDNUjRJ5L7gIBRz24RDY0nSWVIJAmPks5qSlEVgaJpZM4JCjuB
.

littleskunk · 2016-07-05T16:31:23Z

http://stackoverflow.com/questions/8032928/can-i-trigger-javascripts-garbage-collection

littleskunk · 2016-07-06T14:01:13Z

I have to give up. It looks like this code line is the problem:
https://github.com/Storj/storjshare-gui/blob/75b246ac6ac74b3f2082c5f477c714661d91f048/app/lib/logger.js#L48

My memory snapshot is full of log line strings. The garbage collector doesn't delete the old strings and I don't understand the reason for this. Every time this line of code is executed I have one more sting in my memory.

I removed https://github.com/Storj/storjshare-gui/blob/75b246ac6ac74b3f2082c5f477c714661d91f048/app/lib/logger.js#L31-L40 to be sure it is not the array operation. Same result. Memory full of log line strings.

At the moment I see only one solution: #266

RedRockMining · 2016-07-08T05:35:58Z

I'm not sure how #266 was supposed to correct this, but using the latest code (0.8.9) I am now getting (also after an hour or so) a WSD (White Screen of Death). However now, when I select View->Reload I get a javascript error identical to #270 which is independently reproducible without the BSD/WSD error. Prior to this latest version, View->Reload correctly restarted the application.

matanbr · 2016-07-08T08:04:42Z

any chance to make a new binary with loglevel warn for 0.8.9 ?

littleskunk · 2016-07-08T15:01:41Z

@RedRockMining #266 will come with GUI v0.9.0.

@matanbr Give me a few minutes.

littleskunk · 2016-07-09T07:19:58Z

Hotfix:
https://github.com/Storj/storjshare-gui/releases/download/v0.8.9/storjshare-gui.loglevel.info.exe
Same loglevel as the normal GUI release but the reload button is working.

https://github.com/Storj/storjshare-gui/releases/download/v0.8.9/storjshare-gui.loglevel.warn.exe
Reduced logging. This version will crash a few hours later. Watch for Contract no longer open for offer to be sure it is running fine.

https://github.com/Storj/storjshare-gui/releases/download/v0.8.9/storjshare-gui.loglevel.off.exe
No logging. This version should not crash and run for days. You can only look into your shard directory to see if it is running fine.

MeijeSibbel · 2016-07-09T10:37:00Z

What i found so far:

I only get a white-screen on the GUI when storj auto-starts after windows booting.
After manually resetting the GUI no further issues occur.

littleskunk · 2016-07-09T10:55:58Z

@MeijeSibbel that looks like a new issue. I would expect the "white" screen of death after a few hours. If you get it asap it must be something else. If you can reproduce it please open a new issue. Developer console error messages would help.

MeijeSibbel · 2016-07-10T14:46:49Z

@skunk: I added some info to the "white screen" github page. I can reproduce it by restarting my windows 10 machine and letting storjshare pop-up due to the auto-start function, it will pop-up with the white screen every time. After the reset it will not appear again, unless the machine is restarted again. Although the screen is white, it will still download shards. hope this helps.

RedRockMining · 2016-07-11T18:29:34Z

Updated to v0.9.0 this morning just to see if anything had changed. Still getting WSD after about an hour and Javascript error #270. Re-downgraded to loglevel.off, which is working fine. Not sure if this is being worked on at the moment, but please let us know if there is any progress or if the loglevel options might be added to the current release so those of us with this problem can run the latest code.

littleskunk · 2016-07-11T22:37:59Z

@RedRockMining did you notice the advanced options? You can now configurate the loglevel yourself.

RedRockMining · 2016-07-11T22:47:20Z

My apologies, I missed it!. I was assuming if there was something in v0.9.0 to address this issue, there'd have been a comment here but no worries, I'll try 'Show Errors Only' and see how that goes!

Felmane · 2016-07-14T14:35:29Z

This a general view of my gui client. This was taken about 30 minutes after it was running. The memory usage is as You can see about 700mb. 5 minutes later it dropped around 520mb, only to rise up to 600mb 10 minutes later.

It is running on a fresh install of Win10 Pro 64bit. The hardware is an Acer Laptop 5755G (i7-2670QM - 4gb of ram).
Shards folder is empty, farmer.db folder is shown in the screenshot.

There are 4 processes of Electron in the task manager. The first shown in the screenshot with a CPU usage from 12% to 18%., the other 3 use about 0.1% - 0.3% of processor and from 12mb to 36mb of ram. (I guess the extra processes are for the connections? Since im setup to tunnel up to 3 connections)

The WSOD (white screen of death) happens an hour or two after start. It starts automatically at Windows boot and no problems occur at startup unlike what described by @MeijeSibbel .
Once it happened when I lost connection to the internet, the other times I cannot be certain since I would find it after unlocking the laptop (from the win login screen) or by connecting through remote desktop connection. Maybe it is somehow related?

I tried to set the log level to "debug" too, but there doesn't seem to be anything out of the ordinary.
Hope this helps. If you can tell where are the logs stored i can send you those too.

EDIT: It just crashed in front of me. The main process died using a peak of 20% cpu and at about 730mb of ram. The other 3 smaller processes are still running.
The developer tools sent an error message stating it got disconnected from the main page.
Reloading from the Menu restarts the client fine and its running again.

littleskunk · 2016-07-14T18:26:21Z

@Felmane Configurate loglevel off and it will run for days without any crash. https://storj.readme.io/docs/storjshare-troubleshooting-guide#logfile-and-loglevel

super3 added the bug label Jul 1, 2016

littleskunk self-assigned this Jul 5, 2016

161chihuahuas modified the milestone: v0.9.0 Jul 5, 2016

aleitner modified the milestones: v0.9.0, v1.0.0 Jul 7, 2016

littleskunk mentioned this issue Jul 8, 2016

StorjShare - blank #271

Closed

littleskunk mentioned this issue Jul 9, 2016

JS stacktrace #273

Closed

littleskunk changed the title ~~GUI black screen of death~~ GUI white screen of death Jul 9, 2016

littleskunk mentioned this issue Jul 13, 2016

v0.9.1 Windows Still Crashes Frequently #283

Closed

aleitner mentioned this issue Jul 19, 2016

v1.0.0 #286

Merged

161chihuahuas added the needs review label Jul 19, 2016

littleskunk closed this as completed in #286 Jul 20, 2016

littleskunk removed the needs review label Jul 20, 2016

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

GUI white screen of death #257

GUI white screen of death #257

littleskunk commented Jun 30, 2016

matanbr commented Jul 1, 2016

littleskunk commented Jul 1, 2016

mo35 commented Jul 1, 2016 •

edited

Loading

trippypippy commented Jul 2, 2016

Blackwolfen commented Jul 3, 2016

Blackwolfen commented Jul 4, 2016

Blackwolfen commented Jul 4, 2016

ghost commented Jul 4, 2016

trippypippy commented Jul 4, 2016

littleskunk commented Jul 4, 2016

trippypippy commented Jul 4, 2016

RedRockMining commented Jul 4, 2016 •

edited

Loading

byterunner13 commented Jul 4, 2016

trippypippy commented Jul 4, 2016

littleskunk commented Jul 4, 2016 •

edited

Loading

matanbr commented Jul 5, 2016

littleskunk commented Jul 5, 2016

RedRockMining commented Jul 5, 2016 •

edited

Loading

byterunner13 commented Jul 5, 2016

littleskunk commented Jul 5, 2016

littleskunk commented Jul 6, 2016 •

edited

Loading

RedRockMining commented Jul 8, 2016 •

edited

Loading

matanbr commented Jul 8, 2016

littleskunk commented Jul 8, 2016

littleskunk commented Jul 9, 2016

MeijeSibbel commented Jul 9, 2016 •

edited

Loading

littleskunk commented Jul 9, 2016

MeijeSibbel commented Jul 10, 2016 •

edited

Loading

RedRockMining commented Jul 11, 2016 •

edited

Loading

littleskunk commented Jul 11, 2016

RedRockMining commented Jul 11, 2016 •

edited

Loading

Felmane commented Jul 14, 2016 •

edited

Loading

littleskunk commented Jul 14, 2016

GUI white screen of death #257

GUI white screen of death #257

Comments

littleskunk commented Jun 30, 2016

matanbr commented Jul 1, 2016

littleskunk commented Jul 1, 2016

mo35 commented Jul 1, 2016 • edited Loading

trippypippy commented Jul 2, 2016

Blackwolfen commented Jul 3, 2016

Blackwolfen commented Jul 4, 2016

Blackwolfen commented Jul 4, 2016

ghost commented Jul 4, 2016

trippypippy commented Jul 4, 2016

littleskunk commented Jul 4, 2016

trippypippy commented Jul 4, 2016

RedRockMining commented Jul 4, 2016 • edited Loading

byterunner13 commented Jul 4, 2016

trippypippy commented Jul 4, 2016

littleskunk commented Jul 4, 2016 • edited Loading

matanbr commented Jul 5, 2016

littleskunk commented Jul 5, 2016

RedRockMining commented Jul 5, 2016 • edited Loading

byterunner13 commented Jul 5, 2016

littleskunk commented Jul 5, 2016

littleskunk commented Jul 6, 2016 • edited Loading

RedRockMining commented Jul 8, 2016 • edited Loading

matanbr commented Jul 8, 2016

littleskunk commented Jul 8, 2016

littleskunk commented Jul 9, 2016

MeijeSibbel commented Jul 9, 2016 • edited Loading

littleskunk commented Jul 9, 2016

MeijeSibbel commented Jul 10, 2016 • edited Loading

RedRockMining commented Jul 11, 2016 • edited Loading

littleskunk commented Jul 11, 2016

RedRockMining commented Jul 11, 2016 • edited Loading

Felmane commented Jul 14, 2016 • edited Loading

littleskunk commented Jul 14, 2016

mo35 commented Jul 1, 2016 •

edited

Loading

RedRockMining commented Jul 4, 2016 •

edited

Loading

littleskunk commented Jul 4, 2016 •

edited

Loading

RedRockMining commented Jul 5, 2016 •

edited

Loading

littleskunk commented Jul 6, 2016 •

edited

Loading

RedRockMining commented Jul 8, 2016 •

edited

Loading

MeijeSibbel commented Jul 9, 2016 •

edited

Loading

MeijeSibbel commented Jul 10, 2016 •

edited

Loading

RedRockMining commented Jul 11, 2016 •

edited

Loading

RedRockMining commented Jul 11, 2016 •

edited

Loading

Felmane commented Jul 14, 2016 •

edited

Loading