Page 3 of 8 FirstFirst 12345 ... LastLast
Results 21 to 30 of 76

Thread: Welcome to emergency mode!

  1. #21
    Join Date
    Jun 2008
    Location
    San Diego, Ca, USA
    Posts
    10,935
    Blog Entries
    2

    Default Re: Welcome to emergency mode!

    Quote Originally Posted by arvidjaar View Post
    Did you read what I wrote about emergency mode? This will not work currently because status of all failed units is reset when emergency.target is entered.
    Although I suspect you may be referring to some kind of issue, there is probably more to it than what you describe. Are there any links to where the problem is actually described?

    I just rebooted one of my 12.3 to emergency mode and ran the command I described grepping for "failed" and successfully returned the same results I knew existed when I booted to normal mode... a misconfigured Service associated with dnsmasq I haven't gotten around to troubleshoot yet and a couple systemd Units which look like they're supposed to address some kind of systemd data preservation issue like what you describe (amusing though that they themselves are non-functional and reporting as "failed" also in Emergency mode).

    So, until I know better I'll still suggest the commands I recommended since they have worked for me.

    TSU

  2. #22
    Join Date
    Sep 2012
    Posts
    4,976

    Default Re: Welcome to emergency mode!

    Quote Originally Posted by tsu2 View Post
    I just rebooted one of my 12.3 to emergency mode
    How? What exactly did you do?

  3. #23
    Join Date
    Feb 2009
    Location
    Spain
    Posts
    25,547

    Default Re: Welcome to emergency mode!

    On 2013-04-05 06:06, arvidjaar wrote:
    >
    > tsu2;2544661 Wrote:
    >>
    >> I just rebooted one of my 12.3 to emergency mode

    >
    > How? What exactly did you do?


    When I test this situation I create a bad line in fstab, like a mount of
    an inexistent partitition.

    --
    Cheers / Saludos,

    Carlos E. R.
    (from 12.1 x86_64 "Asparagus" at Telcontar)

  4. #24
    Join Date
    Sep 2012
    Posts
    4,976

    Default Re: Welcome to emergency mode!

    Quote Originally Posted by robin_listas View Post
    When I test this situation I create a bad line in fstab, like a mount of
    an inexistent partitition.
    Unfortunately exactly in this case "systemctl --failed" is of little help. The problem is, no unit has failed; filesystem was not mounted due to timeout waiting for device so it did not even attempt to start this service. With journalctl -b you will see something like:
    Code:
    Apr 05 03:52:36 linux-9qg3.site systemd[1]: Job dev-no-such-dev.device/start timed out.
    Apr 05 03:52:36 linux-9qg3.site systemd[1]: Timed out waiting for device dev-no-such-dev.device.
    Apr 05 03:52:36 linux-9qg3.site systemd[1]: Dependency failed for /t.
    Apr 05 03:52:36 linux-9qg3.site systemd[1]: Dependency failed for Local File Systems.
    Apr 05 03:52:36 linux-9qg3.site systemd[1]: Dependency failed for Remote File Systems (Pre).
    Apr 05 03:52:36 linux-9qg3.site systemd[1]: Job remote-fs-pre.target/start failed with result 'dependency'.
    Apr 05 03:52:36 linux-9qg3.site systemd[1]: Job local-fs.target/start failed with result 'dependency'.
    Apr 05 03:52:36 linux-9qg3.site systemd[1]: Triggering OnFailure= dependencies of local-fs.target.

  5. #25
    Join Date
    Feb 2009
    Location
    Spain
    Posts
    25,547

    Default Re: Welcome to emergency mode!

    On 2013-04-05 10:06, arvidjaar wrote:
    >
    > robin_listas;2544699 Wrote:
    >>
    >> When I test this situation I create a bad line in fstab, like a mount
    >> of an inexistent partitition.
    >>

    >
    > Unfortunately exactly in this case "systemctl --failed" is of little
    > help. The problem is, no unit has failed; filesystem was not mounted due
    > to timeout waiting for device so it did not even attempt to start this
    > service. With journalctl -b you will see something like:
    > Code:
    > --------------------
    >
    > Apr 05 03:52:36 linux-9qg3.site systemd[1]: Job dev-no-such-dev.device/start timed out.
    > Apr 05 03:52:36 linux-9qg3.site systemd[1]: Timed out waiting for device dev-no-such-dev.device.
    > Apr 05 03:52:36 linux-9qg3.site systemd[1]: Dependency failed for /t.
    > Apr 05 03:52:36 linux-9qg3.site systemd[1]: Dependency failed for Local File Systems.
    > Apr 05 03:52:36 linux-9qg3.site systemd[1]: Dependency failed for Remote File Systems (Pre).
    > Apr 05 03:52:36 linux-9qg3.site systemd[1]: Job remote-fs-pre.target/start failed with result 'dependency'.
    > Apr 05 03:52:36 linux-9qg3.site systemd[1]: Job local-fs.target/start failed with result 'dependency'.
    > Apr 05 03:52:36 linux-9qg3.site systemd[1]: Triggering OnFailure= dependencies of local-fs.target.
    > --------------------


    I have an open Bugzilla on exactly this situation.

    Bug 782904 - System doesn't print enough information when it drops into
    emergency mode

    State: new.

    Last comment by Frederic Crozat said that the issue would be much
    improved in 195+, not to be backported to 12.2

    I want to test this in 12.3 when I can, and report again.


    Another typical test case would be a failed fsck of a partition, but
    this one I don't know how to emulate. Perhaps by intentionally dd from
    null into the start of the partition to destroy it.


    Those are typical things that happen to people that get dropped suddenly
    to emergency mode. There is no denial that systemv handled the situation
    much better.


    However, I wonder if reporting issues in Bugzilla is of any use anymore.
    I have 48 open bugs - many of the reports are not solved, they are
    simply closed when the affected release goes out of maintenance. A
    percent did get some attention, some none at all.

    This is typical of my country governments bureaucracy: they simply do
    not answer and hope you tire away :-(


    --
    Cheers / Saludos,

    Carlos E. R.
    (from 12.1 x86_64 "Asparagus" at Telcontar)

  6. #26

    Default Re: Welcome to emergency mode!

    The recent correspondence, though interesting, is on the fringes of my understanding, so I need to spend some time digesting its implications for me. Meanwhile, I can report some results.
    Like me, vim wasn't keen to insert nofail into my fstab without a comma, but did so, and my next boot succeeded in reaching the graphical interface. Unfortunately, the doubts of robin_listas subsequently proved justified. I've tried nofail with and without a comma, and the most I can report is that without one it does seem to increase the proportion of successful boots.
    It seems to me that [Esc]aping the splash screen during the boot is more interesting, yielding the following lines towards the end:
    Code:
    Failed to start Show Plymouth Boot Screen.
    See 'systemctl status plymouth-start.service' for details.
    Timed out waiting for device dev-disk-by\x2did-ata\x2dST3160827AS-4MT08NMF\x2part8.device.
    Dependency failed for /windows/E.
    Dependency failed for Local File Systems.
    Dependency failed for Remote File Systems (Pre).
    Welcome to emergency mode!
    I evidently have more work to do, but wonder what that means to you experts. I haven't noticed any problems in practice with any of my windows partitions, least of all E, which I use all the time.
    I wonder if tsu2's comments about graphics, which seem relevant to the first of these lines, could also relate to KMail. My e-mail has become a major problem since upgrading, because after a variable small number of messages the viewing pane becomes fixed on:
    Code:
    Retrieving Folder Contents
    
    Please wait . . .
    The same thing has also occasionally happened with Konqueror (which now also fails to connect to links, showing /var/tmp/kdecache-peterichardavis/krun/.......... in the Location Toolbar). I'll continue my search for answers, but would welcome any more expert guidance.
    Peter

  7. #27
    Join Date
    Sep 2012
    Posts
    4,976

    Default Re: Welcome to emergency mode!

    Quote Originally Posted by peterichardavis View Post
    Code:
    Timed out waiting for device dev-disk-by\x2did-ata\x2dST3160827AS-4MT08NMF\x2part8.device.
    Dependency failed for /windows/E
    Edit /etc/fstab and remove or comment out line with /windows/E. It refers to non-existing device.

  8. #28
    Join Date
    Feb 2009
    Location
    Spain
    Posts
    25,547

    Default Re: Welcome to emergency mode!

    On 2013-04-06 10:06, peterichardavis wrote:
    >
    > The recent correspondence, though interesting, is on the fringes of my
    > understanding, so I need to spend some time digesting its implications
    > for me. Meanwhile, I can report some results.
    > Like me, vim wasn't keen to insert nofail into my fstab without a
    > comma, but did so, and my next boot succeeded in reaching the graphical
    > interface. Unfortunately, the doubts of robin_listas subsequently
    > proved justified. I've tried nofail with and without a comma, and the
    > most I can report is that without one it does seem to increase the
    > proportion of successful boots.


    Allow me to spank you a bit: no harsh feelings intended, just friendly
    advice in terse language ;-)

    Randomly editing fstab without knowing the correct syntax is absurd on
    your part. The line would be, for example, with the proposed change by
    "AlmostAWhisper":

    Code:
    LABEL=c_home   /home    xfs   defaults,nofail   1 2
    That line would be correct, but would be wrong and cause problems to
    you. You should read the manual and read what "nofail" does. Basically
    it tells the system that if the home partition fails to mount, to
    continue as if nothing had happened.

    Then, _if_ the cause of your emergency mode was a faulty home mounting
    (of which we had no proof), your would get a normal boot - but without a
    home partition! You would not be able to login.

    It would basically solve nothing.

    That goes as explanation of my doubts.


    I continue. Randomly editing an fstab line without knowing the syntax
    will break it and make your system not to boot. You have to know what
    exactly to write, or not do it. As we don't know what you changed on it,
    I suggest you copy here the exact contents of the "/etc/fstab" file,
    inside code tags, so that we can tell you what to edit there.


    I see you used vim for your edits. I can feel your pain there, because
    although vim is a very powerful editor, so much so that it is loved by
    "professionals", it is very difficult to use, specially by novices. I
    have been using Linux since 97, and I still don't know how to use it
    (that's an exaggeration: I do know the basics).

    So, if you need to edit a file in text mode or in a terminal, please use
    "joe". It has menu, it has help. It has several flavors: jstar, jmacs,
    jpico. If they are installed, another possibility is mcedit or nano.

    (Suggestion: every time I install openSUSE, I select "mc" to be added to
    the installation. Try it when you can)


    > It seems to me that [Esc]aping the splash screen during the boot is
    > more interesting, yielding the following lines towards the end:


    That was a very good idea. You taught me something :-)

    You have discovered why I do not like splash screens. They should be
    forbidden.

    And you found the problem. See:

    Code:
    
    > --------------------
    >     Failed to start Show Plymouth Boot Screen.
    >   See 'systemctl status plymouth-start.service' for details.
    >   Timed out waiting for device dev-disk-by\x2did-ata\x2dST3160827AS-4MT08NMF\x2part8.device.
    >   Dependency failed for /windows/E.
    >   Dependency failed for Local File Systems.
    >   Dependency failed for Remote File Systems (Pre).
    >   Welcome to emergency mode!
    > --------------------
    It is that "/windows/E" partition that is giving you problems.

    Do what arvidjaar tells you: comment out the line with "/windows/E" on
    it. That's the line that is causing problems.

    And let's hope it is is the only one :-)

    --
    Cheers / Saludos,

    Carlos E. R.
    (from 12.1 x86_64 "Asparagus" at Telcontar)

  9. #29
    Join Date
    Jun 2008
    Location
    San Diego, Ca, USA
    Posts
    10,935
    Blog Entries
    2

    Default Re: Welcome to emergency mode!

    Quote Originally Posted by arvidjaar View Post
    How? What exactly did you do?
    This was an intentional and voluntary action, no problem with the boot sequence on this machine.

    Re: Carlos post #25
    https://forums.opensuse.org/english/...ml#post2544838

    Yes this is all testing on 12.3, systemd version 195 which is a <big> jump up in systemd functionality compared to 12.1 (first bare appearance) and 12.2(mostly boot sequence functionality only, very little more).

    TSU

  10. #30
    Join Date
    Jun 2008
    Location
    San Diego, Ca, USA
    Posts
    10,935
    Blog Entries
    2

    Default Re: Welcome to emergency mode!

    Quote Originally Posted by peterichardavis View Post
    The recent correspondence, though interesting, is on the fringes of my understanding, so I need to spend some time digesting its implications for me. Meanwhile, I can report some results.
    Like me, vim wasn't keen to insert nofail into my fstab without a comma, but did so, and my next boot succeeded in reaching the graphical interface. Unfortunately, the doubts of robin_listas subsequently proved justified. I've tried nofail with and without a comma, and the most I can report is that without one it does seem to increase the proportion of successful boots.
    It seems to me that [Esc]aping the splash screen during the boot is more interesting, yielding the following lines towards the end:
    Code:
    Failed to start Show Plymouth Boot Screen.
    See 'systemctl status plymouth-start.service' for details.
    Timed out waiting for device dev-disk-by\x2did-ata\x2dST3160827AS-4MT08NMF\x2part8.device.
    Dependency failed for /windows/E.
    Dependency failed for Local File Systems.
    Dependency failed for Remote File Systems (Pre).
    Welcome to emergency mode!
    I evidently have more work to do, but wonder what that means to you experts. I haven't noticed any problems in practice with any of my windows partitions, least of all E, which I use all the time.
    I wonder if tsu2's comments about graphics, which seem relevant to the first of these lines, could also relate to KMail. My e-mail has become a major problem since upgrading, because after a variable small number of messages the viewing pane becomes fixed on:
    Code:
    Retrieving Folder Contents
    
    Please wait . . .
    The same thing has also occasionally happened with Konqueror (which now also fails to connect to links, showing /var/tmp/kdecache-peterichardavis/krun/.......... in the Location Toolbar). I'll continue my search for answers, but would welcome any more expert guidance.
    Peter
    What you've posted here almost completely confirms my speculation, you have a problem most likely related to your installed video driver. It is exactly at this point when Plymouth loads where the system first switches over from the driver GRUB is using to the driver configured in the OS.

    So, the usual recommendations and hopefully solutions described in other threads... and these apply mainly to those who are already running the Nouveau driver. If you're running a proprietary nVidia driver, then, there are more issues to consider besides the following...
    1. Configure /etc/systemconfig "NO_KMS" to "yes" (it's likely still on "no" which is default). As described in the Release Notes. Easiest way to do this is to open YAST, open the "etc/sysconfig editor," search for kms and change the setting.

    2. If the KMS option doesn't work, then do the "nomodeset" option.

    More than likely one of the first two above should have fixed your problem. If they don't work, consider trying a proprietary driver, but know that to switch back to nouveau you will have to manually remove the xorg.conf file.

    As for the failled mount, I consider that an unrelated and non-critical issue which can be cleaned up whenever you wish. As others have described, it most likely simply involves editing your fstab.

    HTH,
    TSU

Page 3 of 8 FirstFirst 12345 ... LastLast

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •