Do you have a stack trace or crash ID?

 Issue 403750  has been merged into this issue.

 Issue 404140  has been merged into this issue.

restart, purge, reinstall, use another machine, pray to God, all
possibilities was tried without success.

Atenciosamente,
Fernando J. Cordeiro
perfectron@gmail.com


2014-08-15 15:02 GMT-03:00 <chromium@googlecode.com>:

Sorry for the trouble - there is a longstanding and tricky to solve bug (see issue 22703 and  issue 23837  for details) that has to do with the way package managers on linux update software that can lead to this problem when chrome has been autoupdated by the package manager while it is still running. 

Did you try restarting chrome and trying again as the error message suggests? 

First time the bug was found on Linux, so I started to use Chormium,
instead. But some days ago I found the same problem in my Windows and
Chromium too. Restart/Reinstall, remove .config/google-chrome or /chromium
do not fix it.

Atenciosamente,
Fernando J. Cordeiro
perfectron@gmail.com


2014-08-15 15:14 GMT-03:00 <chromium@googlecode.com>:

Ok, interesting. Have you noticed whether this happens consistently with any extension, or only with certain ones? 

This happens with all extension

All extensions. After reinstall and made login Chrome or Chromium cant install my extentions back again. This happens in both, Linux and Windows. Only in my Chromebook it is working. 

Downgrade to version 34, issues of version 36 fixed:

- Now can sync and install apps and extensions again;
- Now can be set as default web browser again.
- Now hangouts is working again. No more messages of "Google Talk Plugin has crashed".

Definitely, there is something very wrong since version 35 of Chrome. The code that is causing the bug probably was copied to Chromium as well in the latest version, because since the latest update Chromium has the same frustrating bugs.

Extensions are a very, very important part of Google Chrome experience. I hope developers can help us with this soon. Please, tell me which information you guys need to confirm this bug and work on it. I'm available for help!

Confimed:

Downgraded to version 34, issues of version 36 fixed:

- Now can sync and install apps and extensions again;
- Now can be set as default web browser again.

any link to download old version

Yes, here is: http://goo.gl/3P7ySF

Downloaded the 37 beta, and has the same issue on linux. On windows 7 extensions intall ok in the beta 37.0.2062.94 beta-m

Anyone has a downloadable version 34 32bit instead of 64bit for linux (deb)

Here you can find versions from 28 to 37, in 32 or 64 bits:
http://mirror.pcbeta.com/google/chrome/deb/pool/main/g/google-chrome-stable/

And here is a how to lock the chrome version after the downgrade (in portuguese, use a translator):
http://www.diolinux.com.br/2014/05/google-chrome-35-problemas-com-java-e-plugins-ubuntu.html

Version 34 is the better choice.

Same issue on Ubuntu, Chrome Version 37.0.2062.94 (64-bit). Cant install any extension, error: "Could not install package: 'UTILITY_PROCESS_CRASHED_WHILE_TRYING_TO_INSTALL'. Could not install package because a utility process crashed. Try restarting chrome and trying again."

chrome://crashes/ does not log any crashes

Same issue on Ubuntu 14.04 LTS and Chrome Version 37.0.2062.94 (64-bit).
Same issue also with Chromium and Chrome Beta... I tried to purge, reinstall, clear config files.. but didn't work.

This issue is present with Chrome Version 37.0.2062.94, Ubuntu 12.10 32bit.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/469b8770baa97d9d114df4bd97995f60fb9e39fe

commit 469b8770baa97d9d114df4bd97995f60fb9e39fe
Author: ttuttle <ttuttle@chromium.org>
Date: Sat Sep 06 13:12:10 2014

ReadIpHelper: Zero out before calling GetAdaptersAddresses

This is a first-ditch attempt at fixing  http://crbug.com/401655 .

BUG= 401655 

Review URL: https://codereview.chromium.org/539313002

Cr-Commit-Position: refs/heads/master@{#293610}

[modify] https://chromium.googlesource.com/chromium/src.git/+/469b8770baa97d9d114df4bd97995f60fb9e39fe/net/dns/dns_config_service_win.cc

Using chrome 39 (Dev) and the bug still here.

I went back version by version till iI found one that worked.
Finally Version 35.0.1916.114 worked.
Downloaded form here - http://mirror.pcbeta.com/google/chrome/deb/pool/main/g/google-chrome-stable/

Ubuntu 12.04 ( just about to upgrade to 14.04 ) 64bit

still in Version 38.0.2125.101 (64-bit)
Ubuntu 14.04.1 LTS

Hey, this bug is marked as “Needs-Feedback”. Is there anything we users can help with? Is the proposed patch good?

@27 - what patch are you talking about? I don't see any proposed patch in the comments here (it looks like the auto-added in comment 23 was someone accidentally using the wrong bug id on their codereview, because that patch is windows specific). 

Anyhow, I just tried doing a fresh install of Ubuntu 14.04.1 64-bit in a virtual machine and then tried installing both 32-bit and 64-bit Google chrome, version 38.0.2125.104. In both cases, installs of extensions from the webstore seemed to work just fine.

As a random shot in the dark, is there any chance that those of you encountering this bug have App Armor profiles running for chrome? I think you can tell by doing 'sudo apparmor_status' and looking to see if the list of enforcing profiles includes the binaries /opt/google/chrome/chrome and/or /opt/google/chrome/chrome-sandbox. If so, you might try moving those profiles from enforcing to complain mode following the instructions here: https://help.ubuntu.com/community/AppArmor and see if that helps. 

Otherwise, I'd be curious if anyone who can reproduce the problem could try a test with a clean install of Ubuntu in a virtual machine. If the install works ok, then perhaps you could try grabbing a copy of your chrome profile (http://www.chromium.org/user-experience/user-data-directory) and running with it in your virtual machine to see if that affects things.

@28 Just checked Apparmor status on my machine, either /opt/google/chrome/chrome or /opt/google/chrome/chrome-sandbox is in the listed enforcing profiles, but /usr/lib/x86_64-linux-gnu/lightdm-remote-session-freerdp/freerdp-session-wrapper//chromium is included. Is that related to this issue?

@28 Just checked App Armor here and found /usr/lib/lightdm/lightdm/lightdm-guest-session-wrapper//chromium_browser

@29 and @30 - I don't know if either of those could impact things or not. You might try changing those from enforce to complain mode following the directions at https://help.ubuntu.com/community/AppArmor and see if it makes any difference. 

@31 I've disabled both /usr/lib/lightdm/lightdm/lightdm-guest-session-wrapper//chromium_browser and /usr/lib/x86_64-linux-gnu/lightdm-remote-session-freerdp/freerdp-session-wrapper//chromium profiles, reload all profiles and relaunch Chrome, the issue still exists.

Ok, new test. Kernell updated from 3.5 to 3.17.1, now is everything working again :)

Interesting. To anyone who is still experiencing the problem in Ubuntu 14.04.1, can you report what kernel version you are running (run 'uname -a' in a terminal)? 

I managed to install an extension after installing kernel 3.13.0-37 in Linux Mint 17 Qiana (with Ubuntu 14.04 Trusty Tahr repos). I was previously using kernel 3.5.0-17 and couldn't install any extensions (same UTILITY_PROCESS_CRASHED_WHILE_TRYING_TO_INSTALL error message as above)

I'm experiencing the problem in Ubuntu 14.04.1 with kernel 3.8.0-25.

Please @36 try to update your kernel to the last one (3.17)

I am experiencing this issue as well. Restart / purge / reinstall all failed to resolve my problem.
Linux jamal 3.8.0-19-generic #30-Ubuntu SMP Wed May 1 16:36:13 UTC 2013 i686 i686 i686 GNU/Linux

I just uninstalled and installed an older version of chrome (see below) and extensions worked immediately.

Google Chrome: 35.0.1916.114 (Official Build 270117) 
OS: Linux 

Please, try to update your kernel. You'll be able to use the latest version of Google Chrome without problems.

I am experiencing this issue as well. 

> Linux 3.8.0-19-generic (Ubunutu 6B4b)

I'm having this problem when trying to install new Bookmark Manager from Google
in Chrome Version 38.0.2125.111 (64-bit)
OS: linuxmint 3.8.0-19-generic

@41 and @42 - can you try upgrading your kernel and see if that resolves the problem?

Same problem UTILITY_PROCESS_CRASHED_WHILE_TRYING_TO_INSTALL on Ubuntu 12.04 LTS with 38.0.2125.111 (64-bit) chrome 

Kernel is 
3.5.0-18-generic

People, please, try to update the kernel. It's important to prove that is a kernel related bug and remove the status "needs-feedback" 

We had this problem in Xubuntu 12.04. 
Updated the kernel to 3.16 and the problem went away.

Ok, while it sounds like this is technically a regression, I'm going to mark this as Wontfix because there is a reasonable workaround of updating your kernel.

If that causes great hardship for anyone and you want to do the work to figure out what's going on submit patches to fix it, I can provide pointers for where to start looking and code reviews for the patch.

That will create a problem: any new linux user that try to use chrome will have a lot of headache, because default installation of most distros are using old kernels. I think that find what's going wrong and fix it must be a better idea.

How do you update a kernel?

Please, try this:

http://ubuntuhandbook.org/index.php/2014/08/install-upgrade-linux-kernel-3-16/

I had installed a number of extensions before the error occurred. But once
it happens, no extension will install. I even uninstalled/reinstalled
Chrome, and even then I get the same error.

Apparently something gets set once the error occurs. Google, Chromium, or
somebody needs to write a utility program to clean up the bad
settings/files this error leaves behind.
 On Dec 1, 2014 11:44 AM, <chromium@googlecode.com> wrote:

upgrading my kernel to 3.16 fix this issue for me

Yup, updated to 3.16 and fixed the issue (I'm on ubuntu 12.10) Just follow the link above. ^^ cheers

Thank you so so much! Updating my kernel did the trick!

Same issue on ubuntu 12.04 with kernel 3.5
After upgrading kernel 3.11, issue is fixed.

Thanks!

i got the problem with kernel 3.18.3

which version of chrome ? 

it works here with Chrome Version 40.0.2214.91 (64-bit)
and Linux 3.18.0

It works for me too after upgrading to Linux 3.16.0-031600-generic and Chrome 40.0.2214.94 (64 bits)

My verson of Chrome is 40.0.2214.69 beta (64-bit) and I'm on Linux Mint.  I was experiencing this problem even after updating Chrome, so I did what others suggested and updated my kernel to the latest stable (3.18.5).  I rebooted and the problem is now gone! 

Upgrade Ubuntu 12.04.5 LTS kernel 3.5.0-18-generic to kernel 3.13.0-45-generic solves it! http://askubuntu.com/a/527213/26286

Linux mint here, upgraded my kernel to 3.18.5 and it fixed the problem for me.

More than a little bit unfortunate that this makes Chrome 39 and 40 useless on the latest Ubuntu LTS, 14.04, which comes with kernel 3.8.0-23-generic.

I've been trying to track down this:

Could not install package: 'UTILITY_PROCESS_CRASHED_WHILE_TRYING_TO_INSTALL'. Could not install package because a utility process crashed. Try restarting Chrome and trying again.

For ages, I'll try upgrading to 14.10. Commenting here so that anybody googling hopefully finds it and wastes less time than I did.

Running Chrome 40.0.2214.111 (64-bit) on Linux Mint 17.1 without problems.

But I think Chrome/Chromium has it backwards.  Rather than making every
user play kernel catch-up in order to install extensions, older versions of
Chrome/Chromium should be retrofitted so that extensions install in all
recent kernels.

Or at least if a retrofit is too burdensome, on a going-forward basis,
Chrome/Chromium should maintain full compatibility with all future 3.x (and
maybe 4.x) kernels.  Why should Chrome/Chromium's compatibility problems be
inflicted separately on each of its users?

I was getting this install error when trying to load both secure shell and chrome remote desktop extensions on ubuntu (13.04).  I went ahead and loaded successfully from a windows device but these would not show as loaded extensions when on Linux so I upgraded the kernel to 3.16.0-031600-generic as was suggested here.  Secure shell works but when I launch remote desktop I get Loading ... in another tab and that's it.  Anyone have any ideas?

 Issue 460856  has been merged into this issue.

I believe this issue should be fixed. Many developers use Ubuntu 12.04 and 14.04 on their workstations and they may not have privileges to upgrade their kernel on their work machines. This issue maybe forcing them to switch to another browser because they may be missing on to some very important extension.

This issue should be fixed as not everyone can update the kernel(especially in ARM boards, a near-the-same bug making graphics acceleration not working was discovered on Linux 3.4, and needs changing the VMSPLIT to 2G which makes some apps not working)
I can confirm this bug on a compiled-from-scratch Chromium40 build.

System:    Host: server.linuxnet Kernel: 4.0.0-0.rc2.git1.2.fc23.x86_64 x86_64 (64 bit gcc: 5.0.0)
           Desktop: Xfce 4.12.0 (Gtk 2.24.26)
           Distro: Fedora release 21 (Twenty One)
google-chrome-unstable-42.0.2311.22-1.x86_64
   works just   check with that dev as MANY  API keys failed to be authenticated with the new-ish hash challenge  any dev that  fails to update is imo worthless (especially after being pointed out  by usrbase)

Are you kidding?
We will use Ubuntu LTS 12.04.5 until 2017 year.
For now it has Linux 3.13.

Users may want to boycot Google Chrome.
Why not to use stable APIs, I can't understand.

While 3.17 is the first upstream kernel version to have SECCOMP_FILTER_FLAG_TSYNC, Ubuntu's 3.13 kernel series has a backport of it; see https://bugs.launchpad.net/bugs/1379020; this includes the lts-trusty packages mentioned in comment #72.  Ubuntu's 3.5, 3.8, and 3.11 kernels all reached end-of-life at various points in 2014; those should be upgraded to 3.13 regardless of what Chrome does.

However, Ubuntu's 3.2 kernels are supported by Ubuntu until 2017, and any instance of 12.04 that was initially installed as 12.04.0 or 12.04.1 is still running them unless newer packages were manually installed, if I understand correctly; see https://wiki.ubuntu.com/Kernel/LTSEnablementStack.  Those have seccomp-bpf but not SECCOMP_FILTER_FLAG_TSYNC, so this bug might affect them.

@ #74 jld
Great! Thank you.

Ubuntu 12.04.5 with Trusty kernel:
cat /boot/config-3.13.0-46-generic | grep SECCOMP
  CONFIG_HAVE_ARCH_SECCOMP_FILTER=y
  CONFIG_SECCOMP_FILTER=y
  CONFIG_SECCOMP=y

Ironically, Chrome continues to run flawlessly on Debian Wheezy, since its Kernel 3.2 doesn't seem to support seccomp-bpf, and thus the bug is never triggered (at least on my workstation).

Disabling seccomp-bpf on older Kernels would seem to fix this for all affected users.

Of course, it would also remove one layer of defense, and I don't know enough about Chromium's security architecture to make any statements on that tradeoff (renderer sandboxing vs. security fixes in the renderer itself).

Changing title from

"Not able to install any extension or app, getting error UTILITY_PROCESS_CRASHED_WHILE_TRYING_TO_INSTALL'. Could not install package because a utility process crashed. Try restarting chrome and trying again." 

to

"Problems with seccomp-bpf during extension install on older linux kernels"

to reflect new information. 

jln@ - are you the right person to investigate this?

I don't know what's happening here and we'll investigate. To help dissipate some misinformation: kernel without TSYNC are perfectly supported even in the latest Dev channel (M42). (Also even if we did not support kernel without TSYNC, an obscure crash is not how we would do it).

Could this be related to  issue 439795 ?

I've read this thread and unfortunately there is no information for me to act on.

* Can someone who can reproduce this issue provide a crash report (you can see them in chrome://crashes)?

* Or perhaps run from the command-line: " google-chrome --enable-logging=stderr --v=0"

* Since it could be  issue 439795 , could you:
- Report "uname -a"
- try the test program at https://code.google.com/p/chromium/issues/detail?id=439795#c25 as indicated there

I checked the ubuntu-precise kernel git repo for the commits referenced in  issue 439795 .  The 3.2 and 3.13 branches picked up the cause and the fix, and there are no tagged revisions in the respective broken ranges, which I assume means there were no released packages with the bug.  The 3.5, 3.8, and 3.11 branches have neither patch, because they were out of support when the backporting was done.

So I could be wrong, but I think  issue 439795  has never affected Ubuntu… and there are multiple reports from Ubuntu users in this bug.

For people on Ubuntu, getting the running kernel version from /proc/version_signature would be very helpful.

For the original reporter, 3.5.0-25.39 (from 12.10, Quantal) went EOL in May 16, 2014, even though it was a 12.04 LTS backport kernel. To track backport kernels, you have to keep moving it forward. For example, right now on 12.04 (Precise), to run a backport kernel, you'd want to be installing "linux-image-generic-lts-trusty", which gets you to the next LTS's kernel. This is covered, as jld mentioned in c#74, at https://wiki.ubuntu.com/Kernel/LTSEnablementStack

There is one mention of this problem under 3.18, though. I suspect we may have multiple bugs being reported here.

FWIW, I just tried installing Ubuntu 12.04 and 12.04.5 in vmware virtual machines and wasn't able to reproduce the problem. (I don't know enough about the sandbox particulars to know if testing in a virtual machine like this is valid or not). In each case I got the installer from http://releases.ubuntu.com/ and first tried installing chrome and then installing an extension both before and after applying Ubuntu updates. The details were:


Ubuntu 12.04 (x86_64)
chrome version: 41.0.2272.76
worked ok with kernel versions: 3.2.0-23-generic and 3.2.0-77-generic

Ubuntu 12.04.5 (x86_64)
chrome version: 41.0.2272.76
worked ok with kernel versions and 3.13.0-32-generic and 3.13.0-46-generic

Updating the title so that people who have been mislead into thinking non-TSYNC kernels were deprecated immediately understand that there is simply "some unknown bug" hitting some users.

There is nothing for us to investigate on this bug, I'll leave it open an extra day, but I will close it if nothing comes to us.

If anyone runs into an issue where Chrome crashes on startup, please file a bug.

TSYNC is not required by Chromium, and if you think Chromium is crashing because of the lack of TSYNC in your kernel, it's a bug that we want to fix (or it might be a bug in your kernel that your distribution will want to fix).

Hello
I have the same problem on Linuxmint 17.1 i386. I tried using dgb-chromium to get some stacktrace, but it's like a sub-process crashing, so i don't think the following trace will be useful, but who knows !

# Env:
#     LD_LIBRARY_PATH=/usr/lib/chromium-browser:/usr/lib/chromium-browser/libs
#                PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
#            GTK_PATH=
# CHROMIUM_USER_FLAGS=
#      CHROMIUM_FLAGS=--disable-new-tab-first-run --enable-user-scripts --ppapi-flash-path=/usr/lib/pepperflashplugin-nonfree/libpepflashplayer.so --ppapi-flash-version=16.0.0.305  --enable-pinch
/usr/bin/gdb /usr/lib/chromium-browser/chromium-browser -x /tmp/chromiumargs.71xLb9
GNU gdb (Ubuntu 7.7.1-0ubuntu5~14.04.2) 7.7.1
Copyright (C) 2014 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "i686-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from /usr/lib/chromium-browser/chromium-browser...(no debugging symbols found)...done.
(gdb) handle SIG33 pass nostop noprint
Signal        Stop	Print	Pass to program	Description
SIG33         No	No	Yes		Real-time event 33
(gdb) set pagination 0
(gdb) run
Starting program: /usr/lib/chromium-browser/chromium-browser --disable-new-tab-first-run --enable-user-scripts --ppapi-flash-path=/usr/lib/pepperflashplugin-nonfree/libpepflashplayer.so --ppapi-flash-version=16.0.0.305  --enable-pinch --single-process
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/i386-linux-gnu/libthread_db.so.1".
[New Thread 0xaf58eb40 (LWP 19811)]
[New Thread 0xaebffb40 (LWP 19818)]
[New Thread 0xadfd1b40 (LWP 19819)]
[New Thread 0xad5ffb40 (LWP 19820)]
[New Thread 0xad7d0b40 (LWP 19821)]
[New Thread 0xad7afb40 (LWP 19822)]
[New Thread 0xacdfeb40 (LWP 19823)]
[New Thread 0xac344b40 (LWP 19824)]
[New Thread 0xae33eb40 (LWP 19825)]
[New Thread 0xa6954b40 (LWP 19826)]
[New Thread 0xa610eb40 (LWP 19827)]
[New Thread 0xa590db40 (LWP 19828)]
[New Thread 0xa510cb40 (LWP 19829)]
[New Thread 0xa490bb40 (LWP 19830)]
[New Thread 0xa410ab40 (LWP 19831)]
[New Thread 0xa3909b40 (LWP 19832)]
[New Thread 0xa3108b40 (LWP 19833)]
[New Thread 0xa26fbb40 (LWP 19838)]
[New Thread 0xa1efab40 (LWP 19839)]
[New Thread 0xa16f9b40 (LWP 19841)]
[Thread 0xaebffb40 (LWP 19818) exited]
ATTENTION: default value of option force_s3tc_enable overridden by environment.
[19807:19832:0312/112027:ERROR:proxy_service_factory.cc(104)] Cannot use V8 Proxy resolver in single process mode.
Created new window in existing browser session.
[Thread 0xa1efab40 (LWP 19839) exited]
[Thread 0xa26fbb40 (LWP 19838) exited]
[Thread 0xa3909b40 (LWP 19832) exited]
[Thread 0xa410ab40 (LWP 19831) exited]
[Thread 0xa490bb40 (LWP 19830) exited]
[Thread 0xa510cb40 (LWP 19829) exited]
[Thread 0xa590db40 (LWP 19828) exited]
[Thread 0xa610eb40 (LWP 19827) exited]
[Thread 0xa3108b40 (LWP 19833) exited]
[Thread 0xa6954b40 (LWP 19826) exited]
[Thread 0xa16f9b40 (LWP 19841) exited]
[Thread 0xacdfeb40 (LWP 19823) exited]
[Thread 0xac344b40 (LWP 19824) exited]
[Thread 0xadfd1b40 (LWP 19819) exited]
[Thread 0xaf58eb40 (LWP 19811) exited]
[Thread 0xad7afb40 (LWP 19822) exited]
[Thread 0xad7d0b40 (LWP 19821) exited]
[Thread 0xad5ffb40 (LWP 19820) exited]
[Thread 0xb02a88c0 (LWP 19807) exited]
[Inferior 1 (process 19807) exited normally]
(gdb) backtrace
No stack.
(gdb) thread apply all backtrace
(gdb) quit

BR,
Thomas

Any chance you would be willing to try running Chrome?

The advantage with Chrome is that you could choose to enable crash reporting, which would help us debug this if you give us a crash report ID. There isn't anything for us to go with right now.

Otherwise:

1. Can you confirm things work if you start Chromium with --no-sandbox? How about --disable-seccomp-filter-sandbox?
2. Could you try to run your debug Chromium build with: ./chrome --disable-setuid-sandbox --disable-namespace-sandbox? If it succeeds, please report the status you see in chrome://sandbox

Closing for lack of actionable input on the bug.

If someone runs into this, please open a new bug!

happening again on Version 41.0.2272.101 (64-bit)

Please open new bugs instead of commenting on this one. For there to be any hope of us actually being able to fix the problem, please make sure to specify in your bug report:

-Whether you are running "Google Chrome" as distributed by Google from https://www.google.com/chrome/browser/desktop/index.html, or if you are running Chromium  packaged by your linux distribution. (This is a really crucial distinction)

-Answers to the questions in comment #87 here, which I've copied below for your convenience:
1. Can you confirm things work if you start Chromium with --no-sandbox? How about --disable-seccomp-filter-sandbox?
2. Could you try to run your debug Chromium build with: ./chrome --disable-setuid-sandbox --disable-namespace-sandbox? If it succeeds, please report the status you see in chrome://sandbox


To avoid spurious "me too", "still happening" kinds of comments which don't provide any value without the data we've asked for, I'm closing comments on this bug.