Workaround described here: http://ubuntuforums.org/archive/index.php/t-1341721.html

To go off of tyrone's comment, yes this happens, no I don't know why.
Attached is a shell script that I have created to automate what is found in the 
ubuntu-forums.  

If you get the error above, close down chromium and run the script, and then open 
chromium back up and you should be good.

I don't know how to fix this bug, but I guess this workaround works for now... Would 
like to try and get it fixed though, is there any data I can give to contribute to a 
fix?  Thanks!

I have this too, since a few days. 5.0.321.0 (38439) Ubuntu

Same here, 5.0.307.5 dev.  The script by cunninghamlouis resolved the issue, at least 
for now.

I have created a new script, same thing happens, just nicer code, I guess.
I think we should all publish here what our extensions are, maybe its an 
overlapping/malicious extensions bug?

AdThwart - Version: 0.8.5
Codesite enhancements for Chrome - Version: 1.1
Docs PDF/PowerPoint Viewer (by Google) - Version: 1.5.3
GBX - Google Bookmarks Extension for Chrome - Version: 2.3.6
Google Apps Shortcuts - Version: 1.7
Google Quick Scroll - Version: 0.5.6
One Number - Version: 1.2.0.9
RSS Subscription Extension (by Google) - Version: 2.0.3
Send from Gmail (no button) - Version: 1.11.2

I have a few other disabled extensions, but this will do for now.  Lets try to get 
this narrowed down.

chromiumprofile.sh 170 bytes   Download

This seems to happen to me in Ubuntu when you close the browser then right away try to 
open it again.

I have a profile for which this is occurring. The profile got in this state when my hard drive failed 
(although the profile was on a different drive). Here is some output from a debugger:


[15572:15587:79630379798:FATAL:app/sql/connection.cc(287)] Check failed: false. Could not set page size
Backtrace:
	StackTrace::StackTrace() [0xd41ee9]
	logging::LogMessage::~LogMessage() [0xd54985]
	sql::Connection::OpenInternal() [0xdd5b75]
	sql::Connection::Open() [0xdd5c95]
	history::HistoryDatabase::Init() [0x5dd366]
	history::HistoryBackend::InitImpl() [0x5cc6e2]
	history::HistoryBackend::Init() [0x5ccede]
	DispatchToMethod<>() [0x5b7375]
	RunnableMethod<>::Run() [0x5b73b2]
	MessageLoop::RunTask() [0xd560db]
	MessageLoop::DeferOrRunPendingTask() [0xd567f8]
	MessageLoop::DoWork() [0xd56a48]
	base::MessagePumpDefault::Run() [0xd5c863]
	MessageLoop::RunInternal() [0xd570df]
	MessageLoop::RunHandler() [0xd570ff]
	MessageLoop::Run() [0xd571a4]
	base::Thread::Run() [0xd84a91]
	base::Thread::ThreadMain() [0xd84e83]
	ThreadFunc() [0xd67da8]
	start_thread [0x7ffff44763f7]
	0x7ffff2196b4d

Program received signal SIGTRAP, Trace/breakpoint trap.
[Switching to Thread 0x440ac950 (LWP 15587)]
DebugUtil::BreakDebugger () at base/debug_util_posix.cc:259
259	}
(gdb) bt
#0  DebugUtil::BreakDebugger () at base/debug_util_posix.cc:259
#1  0x0000000000d54c15 in ~LogMessage (this=0x440aa9a0, __in_chrg=<value optimized out>) at 
base/logging.cc:571
#2  0x0000000000dd5b75 in sql::Connection::OpenInternal (this=0x7fffe801baf8, file_name=
    "/home/estade/.config/google-chrome/Default/History") at app/sql/connection.cc:287
#3  0x0000000000dd5c95 in sql::Connection::Open (this=0x7fffe801baf8, path=...) at 
app/sql/connection.cc:67
#4  0x00000000005dd366 in history::HistoryDatabase::Init (this=0x7fffe801bad0, history_name=..., 
bookmarks_path=...)
    at chrome/browser/history/history_database.cc:84
#5  0x00000000005cc6e2 in history::HistoryBackend::InitImpl (this=0x4fe8d90) at 
chrome/browser/history/history_backend.cc:506
#6  0x00000000005ccede in history::HistoryBackend::Init (this=0x4fe8d90, force_fail=false)
    at chrome/browser/history/history_backend.cc:222
#7  0x00000000005b7375 in DispatchToMethod<history::HistoryBackend, void 
(history::HistoryBackend::*)(bool), bool> (obj=0x4fe8d90, 

Evan - I'm not sure that this is a bug.  If the History SQL database got corrupted by 
something, then maybe this is what's supposed to happen.  Brett probably wrote this 
code, so could answer more precisely.

seems there ought to be a better recourse than "show an error dialog every time the 
user starts, forever". I fixed it by deleting all the History files in the profile, 
but I suspect many users would be stuck with a permanently disabled history system.

 Issue 39829  has been merged into this issue.

 Issue 34423  has been merged into this issue.

also, the error message is not helpful. The profile does exist and the user does have 
permissions for it.

(No comment was entered for this change.)

I receive the same error messages, but I do have a working History system. The only 
thing I've noticed that is missing is the Search Engines list: there are none.

Markus was encountering this.  We tracked it to that his previous Chrome failed to 
shut down properly, holding the SQLite files locked.

file ~/.config/google-chrome/Default/*|grep SQLite|cut -d: -f1|xargs -n1 -d '\n' sh -c 'echo $0 >&2;sqlite3 "$0" .schema' >/dev/null
It output things along the line of:
/home/markus/.config/google-chrome/Default/Archived History
/home/markus/.config/google-chrome/Default/Cookies
/home/markus/.config/google-chrome/Default/Extension Cookies
/home/markus/.config/google-chrome/Default/History
Error: database is locked
/home/markus/.config/google-chrome/Default/History Index 2010-05
/home/markus/.config/google-chrome/Default/Login Data
Error: database is locked
/home/markus/.config/google-chrome/Default/Thumbnails
Error: database is locked
/home/markus/.config/google-chrome/Default/Web Data
Error: database is locked
 

i have the same problem. with both chrome and chromium.

(No comment was entered for this change.)

 Issue 49426  has been merged into this issue.

Quote from the bug I just merged ( issue 49426 ) which is interesting:

"the Default/Web Data file is corrupt; It has been truncated to 16544 bytes, and the first 0x4010 bytes are all zero".

So it seems there are at least two possible causes at the moment:

- unclean shutdowns leaving the db locked (comment #16) and
- actual corruption

I see both examples listed under this bug but maybe we should we track them separately given that the underlying cause might be different?

I also got "Profile could not be opened correctly" since two weeks and my history was broken. Comment #16 was very useful for tracking the problem down.

file His*:

History:               data
History Index 2010-07: SQLite 3.x database

ls -lha His*:
-rw-r--r--  1 vmk vmk 2,9M 13. Jul 19:18 History
-rw-r--r--  1 vmk vmk  18M 13. Jul 19:18 History Index 2010-07

I deleted both and the problem is for me solved.

Suggestion 1:
Do an integrity check on the files. If there are any errors, try a backuped version or start with a new clean instance.
Suggestion 2:
Display a better error message with the problematic file names. At least dump them to standard error.

Issue 47996 has been merged into this issue.

The issue I just dup'd in includes simple reproduction steps for at least one instance of this problem.

MX until someone grabs this.

 Issue 54039  has been merged into this issue.

 Issue 52270  has been merged into this issue.

I have a trivial way to reproduce this issue.  Simply start Chrome on Linux, browse around a bit, and pull the power supply out of your computer.  When you reboot, your profile will be locked.

jwb: I believe your repro steps also require you to have your profile on NFS.  sqlite is resilient to this kind of failure; unfortunately, Google's NFS servers appear to hold locks even when the client goes away.  As far as I understand it it's a bug in the server.

evan: noted.  Perhaps the bug could be mutated to cover the issues that are actually chromium's fault, namely 

1) the alert appears twice at launch, and it should be once; 
2) the alert doesn't mention that the profile is locked, it just complains vaguely.  At least a link to workarounds or something would be great;
3) there is no documentation of the --diagnostics flag which root-causes the problem for you

 Issue 57075  has been merged into this issue.

In case it helps someone, here's a variant of evan's script from comment 16 that force-breaks the lock on the files, this appears to fix the issue for me after crashes. (I'm also using a home directory on NFS.)  

Make sure that all Chrome processes are killed before running this:

$ file ~/.config/google-chrome/Default/* | grep SQLite | cut -d: -f1 |xargs -n1 -d '\n' sh -c 'sqlite3 "$0" .schema 2>&1 >/dev/null | grep -q "is locked" && { echo "fixing $0"; mv "$0" tmp.$$; cp tmp.$$ "$0"; rm tmp.$$; }'

I have run into this problem several times after my Ubuntu box rebooted. After a bunch of debugging, I determined that the NFS server was holding locks on the sqlite3 files.  (Unfortunately I didn't find this thread until I'd already debugged it.)  My fix was:

cd .config/google-chrome
mv Default Default.bak
cp -r Default.bak Default

This creates copies of the files without the locks.

I consider the current behavior of showing an error dialog every time Chrome starts without a way for the user to fix the problem to be unfortunate.  I think it would be better to at least let the user reset their profile.  Best of all would be to recover as much state as possible.

Updating the title to include the full sentence for searchability and with what appears to be the most common cause. If there are other causes, please open a separate bug and refer to this one.

This also happened to me after the building lost power. The steps in comment 32 worked for me.

 Issue 59200  has been merged into this issue.

This also happened to me when the building lost power. With the profile locked, attempting to sync never finished, and attempting to cancel it hung Chrome. Other profile actions might present similar problems.

Thanks to the above posters for steps to work around the issue.

I just started getting this problem with Google Chrome 8.0.552.224

I have to click ok twice in order to move and continue to use Chorme.

I then go to re-sync and that doesn't work.

I don't have the problem with the RockMelt browser.....

Is this issue only for Ubuntu users? I am experiencing this problem in Windows. Thanks. 

I can confirm this issue exists on gentoo(kernel 2.6.36-r5, nfs-1.1.4) with chromium 10.0.648.127 after a hard reset.

Thanks Evan for the script. If it worked, Chrome could offer the user to do the same thing  automatically. However it doesn't work for me:


chris@fen-desktop2(.metadata)$ file ~/.config/google-chrome/Default/* | grep SQLite | cut -d: -f1 |xargs -n1 -d '\n' sh -c 'sqlite3 "$0" .schema 2>&1 >/dev/null | grep -q "is locked" && { echo "fixing $0"; mv "$0" tmp.$$; cp tmp.$$ "$0"; rm tmp.$$; }'

chris@fen-desktop2(.metadata)$ chromium-browser --diagnostics
Chrome Diagnostics Mode (2011:3:18.14:33:43)
17 available test(s)
...
[FAIL] Web DB
   DB locked by another process

chris@fen-desktop2(.metadata)$ sqlite3 ~/.config/google-chrome/Default/Web\ Data 
SQLite version 3.6.22
Enter ".help" for instructions
Enter SQL statements terminated with a ";"
sqlite> .schema
CREATE TABLE autofill...

So why is Chrome unable to open "Web DB" (is that the same as Web Data?) even though WebData is not locked and opens fine?

Sorry, I didn't realise chromium stores its profile in a different directory, so the script did nothing. I changed it to this and it worked:

file ~/.config/chromium/Default/* | grep SQLite | cut -d: -f1 |xargs -n1 -d '\n' sh -c 'sqlite3 "$0" .schema 2>&1 >/dev/null | grep -q "is locked" && { echo "fixing $0"; mv "$0" tmp.$$; cp tmp.$$ "$0"; rm tmp.$$; }'

Unbelievable, Chrome (W7) is on level 10 now anbd still I have this error...
Going back to Firefox I guess...

If you think Firefox better suits your needs, then by all means switch to that browser. We aim to be the best browsers for everybody, but there will always be users who have specific requirements that are better met by a different product.

Having said all of this, you will probably discover that both Chrome and Firefox do *much* better if you can avoid storing state files on a filesystem that fails to implement full POSIX semantics, and that is shared across multiple logins.

Long before Chrome existed, when I was using Firefox as my daily browser, I made it a point to move all of the Firefox state files to local disk. I had a lot less trouble afterwards.

All network filesystems (NFS, Coda, ...) have unusual semantics and failure modes. It is better not to use them for parts of the filesystem that are expected to provide full POSIX semantics. This particularly means to never ever use them for /tmp (really bad things happen, if you do!), and to avoid them for some of the files in /home. This especially means ~/.config, ~/.gnome*, ~/.cache, ~/.gconf*, ~/.mozilla and probably a few others that I can't think of right now.

None of this means that we shouldn't try to make a best effort to help users who are unfortunate enough to be stuck with a machine, where local storage isn't available. But I would always expect for some problems to remain, if the filesystem is essentially untrustworthy.

I've had hard crashes where Firefox was in the same fubar'd state after the reboot, where I cannot access my bookmarks / history because the files with that information is locked according to NFS.

Still a problem on v 12.0.742.53 beta
The scripts did not work.
Comment 32 fixed the problem...

From the users directory, run this in terminal..

cd .config/google-chrome
mv Default Default.bak
cp -r Default.bak Default

130+ users have reported this issue on "profile not opening correctly" pop up   in 'GoogleFeedback'. Many users have reported that
they get the message when opening chrome .We have been getting reports in  11.0.696.77 version and previous versions as 10.0.648.205.
Users have reported two types of messages :
1)Unable to open your profile correctly
It is possible that some features are not available, verify that the profile exists and you have permission to access its content in reading and writing
2)Your profile cant be used because it is from a newer version of chrome
Sample User Reports:
http://goto.google.com/30533332
http://goto.google.com/37153482
For more user reports refer the cluster URL below :
Cluster URL
http://goto.google.com/1015824

This issue has just occurred for me after upgrading to 14.0.797.0 dev-m. Every time I restart I get a dialogue box stating:

"Your profile could not be opened correctly. Some features may be unavailable. Please check that the profile exists and you have permission to read and write its contents".

chrome://history doesn't load, and I was signed out of browser sync. I signed back into sync, restarted and I was once again signed out of sync. I tried this one more time and same issue.

I might have to create a new profile as all the suggestions here appear to be relating to fixing this issue using Ubuntu - any ideas on how to fix it in Windows?

I'm on Windows 7 Professional x64.

I have same problem and everything you say means nothing to me.  Why can't Google just fix the problems when I sign on.  I am a dummy when it comes to fixing my own programs.

I solved the problem using the method that everyone is mentioning.

cd .config/google-chrome
mv Default Default.bak
cp -r Default.bak Default

Note that, you have to do the above steps while your browser is closed. Otherwise, it won't work.

i used to have this a few months ago, then something i've found on the net solved it without deleting my profile.

now i'm seeing this again with "Google Chrome 13.0.782.107 beta"

my history is fully functional

no outstanding sqlite locks

profile stored on a local ext4 filesystem

power outages (kernel crashes) might have happened, although definitely rarely

on another note: my Default directory is 1 GB! the overwhelming majority of it is the one year history i have. that doesn't look sustainable over the upcoming years...

I have had this issue many times and they seem to happen almost exclusively when I attempt to create a custom unity launcher for Chromium or when I launch Chromium from terminal. Deleting the profile folder Default in .config/chromium solves the problem but reverts the browser to a freshly installed state.

15.0.853.0 (Developer Build 96745 Linux) Ubuntu 11.04

I have the same problem with Mac OS X 10.6.8. Anyone who knows what I can do to fix this?

it is a problem related to the SQLite database
best solution = delete the profile and resync

How can that be the "best solution"?

Why does this even happen? Why doesn't the NFS server clean up locks when it reboots, or when the client reboots?

just found a weird solution atleast for me. I killed all the chormium processes. I still had some hanging after I closed the browser. Killed all the hanging processes. Restarted the browser, resynced. viola fixed. dont know if this will help anyone but there it is.

14.0.835.202 (Developer Build 103287 Linux) Ubuntu 11.04

It would be helpful if I knew where to go in order to do the steps in comments 32, 48 & 57. So far inputing the command in RUN isn't making it pop up.

I don't see the problem anymore

(No comment was entered for this change.)

The tech talk does not help lay people. I had Chrome installed and could not show my extensions. The suggestion was to uninstall and reinstall. I did and now I get that error about my profile. Can someone tell me what to do in simple English steps?

It is not happening in Chrome dev 17

I got this problem since i tried to install "HOT SPOT SHIELD" to get around a proxy here. 

Now i get that message "Your profile could not be loaded bla bla"

and when I quit Chrome it doesnt quit and has to be FORCED every time. Anyone got an answer yet for this? 

(No comment was entered for this change.)