init: ureadahead-other main process terminated with status 4

Also same here.

Same on AMD64 - it's like this for at least a weak, but I did not have the time to report.

I am also getting the same message. See my entry https://answers.launchpad.net/ubuntu/+question/89741.

This is a normal exit code to indicate that there is no pack file for a separate filesystem you have in your fstab.

The bug is that Upstart displays them (and is already filed elsewhere)

Scott, can you be more specific about the bug so that I can track it (where is it filed)? Thanks.

I had the same (or at least similar) problem. On my computer (Dell laptop D630, intel graphic) this error occurs at every boot. I get a black screen with a white (non-flashing) underscore in the top-left so that I cannot log in any more -- also the text/console login does not work and even the "failsafe" mode fails. If I add the kernel option "i915.modeset=0" in GRUB to the "failsafe" mode, I still cannot log in after boot but I can see the start-up messages, where the last line is "init: ureadahead-other main process (762) terminated with status 4". Hence, I was really locked out from my own computer :-( So, I had to re-install Kubuntu 9.10 and everything runs fine now. Since this issue and a few other problems occurred after upgrading from 9.04 (Jaunty) to 9.10 (Karmic), I think that it is caused by a bug in the (dist-)upgrade procedure.

BTW: I originally installed Ubuntu 8.10 (Intrepid) on this computer, then changed to Kubuntu and then upgraded to 9.04 (Jaunty). All this worked fine but I mention it here, because the upgrade to 9.10 (Karmic) might be affected by the previous upgrade. I have the non-working system still installed on another partition. So please let me know if you have any questions.

I reported this problem also as comments to bug #483359 and bug #483429.

I got a similar one!

Already described everything here
http://ubuntuforums.org/showthread.php?t=1343648

In short terms:
- I can't login anymore (black screen, non-blinking cursor on the upper left, no access to tty1-6).
- I have an intel graphic.
- And maybe interesting: My /home is on an nfs drive. But just /home!
- And I also updated my Ubuntu from 8.10 to 9.04 and then to 9.10.
- The day before this happend the first time (2. Dec 2009), Ubuntu installed that sreadahead and ureadahead packages. So the first restart after that updates (the next day), the problem accoured the first time.

Normally I'm using KDE and kdm (but didn't installed Kubuntu, just installed kde using synaptic). So after accessing the PC using ssh and changing from kdm to gdm everything was fine again.
Did anybody else also see this behavior (specially that kdm - gdm thing)?

Thanks!
kolAflash

Posted some stuff from my xorg.log on
http://ubuntuforums.org/showthread.php?t=1343648

Ugh. I just got hit with this on a recent update that upgraded ureadahead. I get the error message with a blinking cursor and I can't do anything. ctrl+alt+1 or any other number gives me a blank screen with a blinking cursor. I cannot log in at all...

I have this shown during bootup... multiple times...

Like

urreadahead main process terminated with status 5

and

ureadahead-other main process (898) terminated with status 4

with multiple entries with different process IDs...

But there is no problem in booting or anything... machine is working fine...

I have similar problem.

Got these messages during startup:
.....
init: ureadahead-other main process (xxxx) terminated with status 4
.....
init: network-interface (lo) pre-start process (xxxx) terminated with status 1
init: ureadahead-other main process (xxxx) terminated with status 4
init: ureadahead-other main process (xxxx) terminated with status 4
.....

then starts gdm login screen and I can log in without problem.

But...

I have 2 encripted partitions (swap, priv_data mounting in ~/priv_data) and second - priv_data - does not wait for password.
So if I want to mount this I must switch to text console and type password few times (sometimes it prints my password).
tty1-6 are waiting for mounts
Bluetooth does not start even if it is enabled in /etc/rc[2345].d/ and I have to start it manualy (it's annoying bcs I have bt mouse).

Don't know if my problems are related to this bug, but I couldn't find any other source of these problems.

I get the same messages (exit status 4) 2 times - boot continues after that. I'm on a fresh install of karmic (relates to #7) with four partitions (ubuntu system, home, a data partition, and swap). I have a 32bit intel Core2Duo.

Scott, it would be nice to get some more information about it. As far as I understand ureadahead, it should generate the pack files itself when running the first time?

Hi,

to everybody experiencing those nasty exit statuses. I'm almost sure the problem is described in Bug #432360. If the output of "ureadahead --dump" is that there is no pack file on your system, the solution is in that report.

I have seen the same message of "ureadahead....terminated with status 4" since about a month or less. This is karmik 64bit with dual gnome and kde installed as new (not upgrade). Before there was no message like this. It appeared after an update (damn those screwed updates). Although I have no problems at all (or seo it seems), I believe that message is not normal. In fact, this is the first time since I use linux for 9 straight years. Hope this bug will be solved in a new good update.

This message never goes away for me no matter how many times I reboot and all of the "fixes" I've tried do not work. No pack files ever get generated in /var/lib/ureadahead/ (other than the debugfs directory). I have / mounted on one partition and /home mounted on another. ureadahead doesn't appear to be working on karmic at all.

I assume this is why ureadahead isn't:

# ureadahead -v
/var/lib/ureadahead/pack: No such file or directory
ureadahead: Error while tracing: No such file or directory

And I assume the "error while tracing" is because the -updates kernel isn't installed on my system since the kernel version is pinned. I've never tried to update the kernel to -updates, and I can't seem to find a really clear set of instructions on doing that (like, what about modules?) so I'll probably skip it.

I had this problem too and I found the solution here:
http://tech--help.blogspot.com/2009/12/ubuntu-solved-ureadahead-other-main.html

I had a Windows partition mounted in /etc/fstab that I commented out and now I do not have this error anymore.

I believe the problem was that I hibernated Windows and mouting was failing all the time in a loop. Comment the windows mounting line as said above and all will back to normality.

On Mon, 2009-12-28 at 18:01 +0000, BigBaaadBob wrote:

> And I assume the "error while tracing" is because the -updates kernel
> isn't installed on my system since the kernel version is pinned. I've
> never tried to update the kernel to -updates, and I can't seem to find a
> really clear set of instructions on doing that (like, what about
> modules?) so I'll probably skip it.
>
If you installed ureadahead from -updates, you need the kernel from
-updates too.

Scott
--
Scott James Remnant
<email address hidden>

During Ubuntu 09.10 boot, the message "init: ureadahead-other main
process (473) terminated with status 4" also appears on my system.

/etc/fstab is as follows:
     /dev/sda3 / ext4 errors=remount-ro 0 1
     /dev/sda6 /d ext3 defaults 0 2
     /dev/sda5 swap swap sw 0 0
     tmpfs /tmp tmpfs defaults 0 0
Note /dev/sda6 is a data partition. The rest is self-explanatory.

/var/lib/ureadahead contains TWO pack files:
     "pack" (presumably for the root file system);
     "tmp.pack" (presumably for the /tmp file system).
/var/lib/ureadahead does NOT contain a pack file named "d.pack".

Ubuntu Bug 432360 contains the following text:
"ureadahead has better support for separate filesystems, when
 tracing it will generate additional pack files for each of the
 separate mounts (each one individually optimised) - and uses
 the mountall-generated 'mount' event to run over those."

How can I investigate why ureadahead generates pack files for
the root and /tmp file systems, but not for the /d file system?

I upgraded (using update-manager) to Ubuntu 10.04 Alpha 2. The message is
still there.

On Tue, Jan 19, 2010 at 15:37, Bob Hill <email address hidden> wrote:

> During Ubuntu 09.10 boot, the message "init: ureadahead-other main
> process (473) terminated with status 4" also appears on my system.
>
> /etc/fstab is as follows:
> /dev/sda3 / ext4 errors=remount-ro 0 1
> /dev/sda6 /d ext3 defaults 0 2
> /dev/sda5 swap swap sw 0 0
> tmpfs /tmp tmpfs defaults 0 0
> Note /dev/sda6 is a data partition. The rest is self-explanatory.
>
> /var/lib/ureadahead contains TWO pack files:
> "pack" (presumably for the root file system);
> "tmp.pack" (presumably for the /tmp file system).
> /var/lib/ureadahead does NOT contain a pack file named "d.pack".
>
> Ubuntu Bug 432360 contains the following text:
> "ureadahead has better support for separate filesystems, when
> tracing it will generate additional pack files for each of the
> separate mounts (each one individually optimised) - and uses
> the mountall-generated 'mount' event to run over those."
>
> How can I investigate why ureadahead generates pack files for
> the root and /tmp file systems, but not for the /d file system?
>
> --
> init: ureadahead-other main process terminated with status 4
> https://bugs.launchpad.net/bugs/484677
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Comment #5 states:
"This is a normal exit code to indicate that there is no pack file
 for a separate filesystem you have in your fstab. The bug is that
 Upstart displays them (and is already filed elsewhere)."

Where is this Upstart bug filed (so that I can track it) ?

Scott says: "If you installed ureadahead from -updates, you need the kernel from -updates too."

The problem is that things from -updates just get installed, but the kernel is pinned for any one release of Ubuntu and updating it would require specific user actions that are not (well) documented anywhere. As a result you get this version skew that apparently is still not resolved in 10 Alpha (according to Ernst).

I disagree with the claim that Upstart is wrong to display these errors: while it might be true that this is normal functioning for ureadahead, it is NOT normal functioning for the configuration in which ureadahead is being used. IOW, the boot speedup SYSTEM (using ureadahead) is not working, and Upstart correctly display error messages about that.

I am experiencing this problem. 10.04 A2 amd64 takes ~40 seconds to boot. 9.04 amd64 took around 45 seconds to boot. /var/lib/ureadahead/pack does not exist on my laptop.

i have same problem on i386 ubuntu 10.04 alpha 2 after yesterday dist-upgrade.
system can't boot

I'm seeing this on an Atom 330 processor on an 10.04 upgrade.

It stops at the ureadahead line and doesn't continue any further.

I will boot a rescue CD and look at /etc/fstab for ideas.

Ah, my actual problem was caused by bad /etc/fstab entries referring to disks that were not attached; after commenting them out it boots correctly.

present too in lucid alpha 2

If you add "profile" to your kernel params, ureadahead will create pack files for the mounted volumes. These errors do not show on the subsequent boot does but it does on following boots.

About my fstab: I got three entries which are not mounted during startup,
but are mounted automatically using this script:
http://ubuntuforums.org/showthread.php?t=430312&highlight=sshfs+mount+internet
Actually, two are sshfs and one is a samba share mounted using cifs. I guess
those entries are causing the errors here. Furtermore, I got three ntfs-3g
shares, one swap and two ext4 (/home and /).

Which other bug reports are related to this one? I see several people
mentioning other threads, but no one is actually marked as the actual source
of this problem.

On Thu, Jan 28, 2010 at 17:46, jordanwb <email address hidden> wrote:

> If you add "profile" to your kernel params, ureadahead will create pack
> files for the mounted volumes. These errors do not show on the
> subsequent boot does but it does on following boots.
>
> --
> init: ureadahead-other main process terminated with status 4
> https://bugs.launchpad.net/bugs/484677
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Seeing the same issue on 10.04 A2 under VMware ESX but noticed a grub error "no such disk" on the way there. Now getting "mountall: fsck /boot [632] terminated with status 1".

I have the same issue (Linux kernel 2.6.31-17-generic x86_64 GNU/Linux) when mounting two NTFS partitions (Windows XP). When I uncomment the two lines in /etc/fstab, there is no error. Then only the /, swap and cdrom are mounted.

I have an external USB-HDD with 2 NTFS partitions, that are mentioned in my /etc/fstab.
However, if I turn the HDD off, my system won't boot and stops with the obnoxious statement mentioned by Ernst in Post 1. If I turn on my external HDD, system boots normally, without any error. For me, the error only shows up if ureadahead has no access to these 2 partitions from /etc/fstab.

Alessandro: your system may not boot because you don't have "noauto" as a mount option.

I get the "status 4" error with a nfs mount that is set to noauto:

192.168.1.2:/home/jordanwb /home/jordanwb/RemoteJordan nfs noauto,noatime,user 0 0

Although the error message "init: ureadahead-other main process terminated with status 4" seems to be not related to errors that prevents one from logging in, this error message is still shown on many systems (on the 8th console, Ctrl-Tab-F8). It is shown on my system although all file systems are on a single internal hard disk. Hence, I don't think that these bug reports are invalid.

They're Invalid ureadahead bugs because it's not ureadahead printing that message

So who is the culprit that prints this error message? Why is there an error message about ureadahead? Is there a problem with ureadahead or not?

Scott: I don't understand the logic. Ureadahead is NOT operating properly (which is evident from all the reports here) but because it isn't the one printing the message this bug is invalid?

Ureadahead NEVER works on my system which has, as I noted above, three mounts: /, /boot, and /home. It never works on lots of other systems as reported above.

There needs to be a bug report that tracks the problem of Ureadahead not working on all these systems, nomatter what software is reporting the problem. Or it needs to be removed from Ubuntu.

On Mon, 2010-02-15 at 15:18 +0000, BigBaaadBob wrote:

> Scott: I don't understand the logic. Ureadahead is NOT operating
> properly (which is evident from all the reports here) but because it
> isn't the one printing the message this bug is invalid?
>
There is only one report here, from Ernst filed on 2009-11-18 in which
the computer booted normally and ureadahead worked, other than a message
on the console from Upstart.

The fact that other users, including yourself, have bundled in with
different problems is irrelevant. This bug is just about a harmless
message when you have other filesystems listed in your /etc/fstab

> Ureadahead NEVER works on my system which has, as I noted above, three
> mounts: /, /boot, and /home. It never works on lots of other systems as
> reported above.
>
> There needs to be a bug report that tracks the problem of Ureadahead not
> working on all these systems, nomatter what software is reporting the
> problem. Or it needs to be removed from Ubuntu.
>
I don't disagree. If ureadahead is not working, then you need to file a
bug.

A new bug.

Then we'll debug and diagnose it there.

Scott
--
Scott James Remnant
<email address hidden>

But if there is nothing wrong with ureadahead, no such message should have been shown in the first place. I suspect that ureadahead is not working due to the fact that 10.04 generally boots slower than 9.04.

why invalid? i hate when bugs are invalid, and they're so valid for me. what does that mean?

this happens to me too since the first releases of lucid lynx until today. i'm updating daily.

It's really really a bug! Without read ahead,boot time is more than 60s.I tried many times to make it work,but failed. Last time I boot ubuntu 9.10 in recovery mode,then i had pack files,and boot time is less than 40s.

Given that so many people report and confirm this bug, I have changed the status to "Confirmed".

This message appears by default on all boots of simple ubuntu server installs of A3.

Since it seems mounts are relevant here:
/dev/mapper/ubuntu-root / ext4 rw...
/dev/sda4 /boot ext2 rw...

This is the default for automatic partitioning using LVM on a 1.5 GB disk.

I always had this issue with Karmic. Now I looked in /etc/fstab to see if I can find anything out of ordinary there and found this:
/dev/scd0 /media/cdrom0 udf,iso9660 user,noauto,exec,utf8 0 0

It seems to be normal. At least I know that that was put there during install automatically, because I never ever touched the /etc/fstab file.

But, shouldn't the optical drives be handled by some automounting system? Then why is that entry in the fstab file? I tought I'll give a try and commented that line out to see what happens. Tada! Now when my machine boots up I do not see those ureadahead messages (altough I do see the fsck messages - these messages were always visible along with the ureadahead messages), and my DVD drive also haven't stopped working.

Can anyone give an insight on what's this about?

I don't have that in my fstab file.

Scott,

Why do you change the status back to "Invalid"? Either the error message is true (then there is a bug in ureadahead) or the error message is wrong (then there is a bug in the reporting system). How can there be an error message without an error?

On Thu, 2010-02-25 at 18:47 +0000, Oernie wrote:

> Why do you change the status back to "Invalid"? Either the error message
> is true (then there is a bug in ureadahead) or the error message is
> wrong (then there is a bug in the reporting system). How can there be an
> error message without an error?
>
AS I HAVE SAID OVER AND OVER AGAIN IN THIS BUG REPORT, THE MESSAGE IS
JUST A WARNING FROM UPSTART THAT SHOULD NOT APPEAR ON THE CONSOLE -
THERE IS ALREADY A BUG FILED ON UPSTART THAT THE WARNING IS VISIBLE AT
ALL.

NOW, IF YOU ARE HAVING SOME KIND OF PROBLEM WITH UREADAHEAD IT IS
*NOTHING* TO DO WITH THIS MESSAGE AND NOTHING TO DO WITH THIS BUG!

FILE YOUR OWN BUG! STOP TRYING TO HIJACK THIS ONE!

Scott
--
Scott James Remnant
<email address hidden>

Caps lock. Are you ready to unleash the f&&king furry?

Comment #5 states:
"This is a normal exit code to indicate that there is no pack file
 for a separate filesystem you have in your fstab. The bug is that
 Upstart displays them (and is already filed elsewhere)."

Comment #50 states:
"The message is just a warning from upstart that should not
appear on the console - there is already a bug filed on upstart
that the warning is visible at all."

Where is this Upstart bug filed (so that I can track it) ?

Bob,
I totally agree with you!

Scott,
If this bug report is a duplicate, then set the status to "Duplicate", *not* "Invalid".

On Fri, 2010-02-26 at 07:34 +0000, Bob Hill wrote:
> Comment #5 states:
> "This is a normal exit code to indicate that there is no pack file
> for a separate filesystem you have in your fstab. The bug is that
> Upstart displays them (and is already filed elsewhere)."
>
> Comment #50 states:
> "The message is just a warning from upstart that should not
> appear on the console - there is already a bug filed on upstart
> that the warning is visible at all."
>
> Where is this Upstart bug filed (so that I can track it) ?
>
In the Upstart *upstream* (not Ubuntu) bug tracker, it's title has
something to do with "normal exit"

Scott
--
Scott James Remnant
<email address hidden>

It's this bug: https://bugs.launchpad.net/upstart/+bug/522197

On Fri, Feb 26, 2010 at 17:15, Scott James Remnant <email address hidden>wrote:

> normal exi

To Botond Szász:

I have identical input line in /etc/fstab

/dev/scd0 /media/cdrom0 udf,iso9660 user,noauto,exec,utf8 0 0

I uncommented this line, and reboot. The problem is still there. ureadahead ...terminated with statuts 4 (or 5), in several instances.
It seems that fdisk check at boot is not working properly (or it starts, then is killed without checking the partitions). I wonder if this is caused by ureadahead. So, I confirm this bug in Karmic 64bit.

I have this message in Lucid. Also during this error Ubuntu can't mountall.

Beta-1 still have this problem. I can't use Lucid on my MSI vr321x laptop - it's freezes on boot.

I cant install Beta-1 from CD on my machine which has inbuilt Via Unichome.

I have this problem with lucid beta-1: freezes on boot 80% of the times. After 4-5 reboots it will get to GDM.

I have ubuntu lucid x64 with latest updates and nvidia proprietary driver, and i this is the beautiful splash screen i see ^

is this resovled? my box can not boot!!!Not even single user mode!

My box is dell precision 340, I am jumping upgrading from 8.04 to 10.04 with ubuntu i686

the system hangs after a full screen of udev warnings(after I tried my method 3 in the following, before is the same as Daniel Añez Scott 's screen shot! )

my system is upto today's upstart package and ureadahead package.
I have separated /boot / /tmp on different partions as
sda1 /boot
sda2 swap
sda3 /tmp
sda5 /
sda6 /home

I tried the following, none of them work

1,change /etc/ureadahead.con and made it mount on localfilesystem
2, disable non critical partitions in /etc/fstab while only keep /boot / partitions
3 change /etc/ureadahead.conf to normal exit with status 0 and 4

HI,

I have same problem with Ubuntu 10.04 lucid lynx RC

I see this problem on my machine after upgrading from 9.x to 10.04 rc 1. My machine no longer boots after this upgrade - stops after this message.

@Everyone
Please try to refrain from commenting on duplicate bug reports. As the little box below states, "Comment here only if you think the duplicate status is wrong." Otherwise, discussion on this bug should be done on the master bug (bug 522197). Thanks in advance.

@Kevin
Boot failures are not related to this bug. As Scott said in comment #40, "There is only one report here, from Ernst filed on 2009-11-18 in which the computer booted normally and ureadahead worked, other than a message on the console from Upstart." Many people get this bug and can boot fine (including myself). Thus your boot problems (and everyone else's for that matter) are most likely completely unrelated to this bug. Please file another bug report for your boot issues, if one hasn't been filed already. Thanks.

See the "duplicate" bug report for one solution.

Its been i have upgraded to the new dist lucid lynx and then i was doing fine.

Today morning i restarted my laptop as the wirless wasnt connecting and i saw the boot hung after displaying this message and i waited waited and still waiting for the laptop to boot..

IT is not..!! I have made any recovery cds or boot up disks.

Need help to fix it

Thank you

Same Problem here:
New installation from Ubuntu Server 10.04 x64 on a Dell PowerEdge 2950.

Gabe (post #64), thank you for finally clarifying what one should do when one finds this page. Your instructions should be moved to the top of the post for all of us average users of Ubuntu who find this page via Google. That is the primary reason why I am posting at this duplicate page.

The second reason I am posting here is because this is the first page that shows up when I Google "init: ureadahead main process terminated with status 5", and Scott James remnant's behavior on this page deserves some scolding. He seems arrogantly dismissive of anyone who is not a thoroughly experienced *nix user, and therefore he does not belong in the Ubuntu community. His snobbish dismissal of other peoples' confusion does not belong in a distro intended for busy people trying to get other work done, people who are puzzled by weird messages when they boot. Thank you Ernst (post #55) for finally including the Bug number of the problem that Scott repeatedly alluded to without actually helping us.

Indeed, the ureadahead warning message may not be related to boot problems. However, if the boot failure happens shortly after the prompt occurs, it is the last thing users will see on-screen, and thus we will suspect that it has something to do with boot failure, simply because some process posted it to the screen before the boot process hangs. So us normal users should be advised to look elsewhere, as Gabe has finally done.

This ureadahead error message emerged for me with the update to 10.10. I will chase it down at https://bugs.launchpad.net/upstart/+bug/522197, and hope that the older fixes are still relevant. For other, mortal users, here is another thing that may help: on my netbook, certain processes will stop and wait unless there is some mouse or keyboard input. I have found that the safest input to get things booting again is to press the <shift> key; wiggling the mouse or touching the trackpad may cause the cursor to freeze once the desktop manager loads up (GDM, at least).