Plesk is not accessible: Can't open or create shared memory by shm.name

Follow

Comments

40 comments

  • Avatar
    Julian Fürderer

    Hello,

    this problem occured today, but i cant try this an the following solutions

     

    Unable to open Plesk: Lock Manager error: Can't open or create shared memory by shm.name

    or

    Plesk is unavailable after upgrade: Can't open or create shared memory by shm.name


    because i can currently only connect to Server via SSH when in rescue mode. (Dont know why ... idea how to enable/check SSH, so i can connect in normal mode ??)

    How can i fix the problem when the rescue mode is running?

  • Avatar
    Unknown User

    Hey is there a solution when i don't have ssh access anymore?? Only the in system rescue... Thanks!

  • Avatar
    Alexandr Redikultsev

    Hi, @Sean Unett and @Julian Fürderer.

    In rescue mode, create a cron task under root user to change the ownership and permissions, like this for example:

    */1 * * * * mkdir -p /run/lock/pmlib && chown root:lock-manager /run/lock/lmlib && chmod -R 0770 /run/lock/lmlib

    Then reboot the server and wait for some time, cron task should do the trick.

  • Avatar
    Pera Burek (Edited )

    VPS Server hosted by Strato.de - Strato Support Team refuses to support Plesk

    Plesk Support Team refuses to support Plesk-key-by-Strato, playing ping-pong with user

     

    server is not accessible for days, because not only Plesk admin interface is broken, but SSH is blocked by Plesk for unknown reason

     

    here is output from Strato Rescue system:

     

    Welcome to Ubuntu 16.04.2 LTS (GNU/Linux 4.4.0-042stab134.3 x86_64)

    * Documentation: https://help.ubuntu.com
    * Management: https://landscape.canonical.com
    * Support: https://ubuntu.com/advantage
    =============== VPS REPAIR MODE ===============
    Your VPS is under repair mode now.
    You can access the file system of your VPS under /repair directory.
    To finish repair mode login into the Customer Service Area
    (https://config.stratoserver.net) and select
    'Server Configuration' -> 'RecoveryManager'.
    =============== VPS REPAIR MODE ===============
    Last login: Mon Nov 26 11:30:59 2018 from -- cut --
    root@h2669559:~#


    when I run Plesk "solution" it causes chown error

    root@h2669559:~# mkdir -p /run/lock/pmlib && chown root:lock-manager /run/lock/lmlib && chmod -R 0770 /run/lock/lmlib
    chown: invalid group: 'root:lock-manager'
    root@h2669559:~#


    if I try with "/repair" in front, it causes chown error

    root@h2669559:~# mkdir -p /repair/run/lock/pmlib && chown root:lock-manager /repair/run/lock/lmlib && chmod -R 0770 /repair/run/lock/lmlib
    chown: invalid group: 'root:lock-manager'
    root@h2669559:~#

     

    here is stat output

    root@h2669559:~# stat /repair/run/lock/lmlib
    File: '/repair/run/lock/lmlib'
    Size: 4096 Blocks: 8 IO Block: 4096 directory
    Device: 9157h/37207d Inode: 499647313 Links: 2
    Access: (0770/drwxrwx---) Uid: ( 0/ root) Gid: ( 100/ users)
    Access: 2018-11-23 13:18:54.222952461 +0100
    Modify: 2018-11-23 13:18:54.222952461 +0100
    Change: 2018-11-23 13:38:57.523406743 +0100
    Birth: -
    root@h2669559:~#
     

     

    Solution suggested by @Alexandr Redikultsev doesn't work

    Do you have solution for this problem?

  • Avatar
    Unknown User

    Hi @Alexandr Redikultsev,

    Thanks for the reaction...

    I tried your solution! I edditing the /repair/etc/contab file and added the line you provided. But when i restart my server it still gets the same error.

    I also tried instead of run. var in the line! But it doesnt do it...

  • Avatar
    Claus Siebeneicher

    Seems as if plesk also uses other files in the /run/lock/ directory if plesk's "Enhanced security mode" is enabled, which could cause issues with (plesk) firewall changes, ssh(d) etc.,  right? 

  • Avatar
    Ivan Postnikov

    Hello @Pera

    Thank you for the feedback.

    As I can see the issue now is resolved with the assistance of Plesk support:

    1. To make Plesk accessible engineer has added corn task manually via rescue mode.
    The path: repair/var/spool/cron/crontabs/root
    The task: 

    @reboot mkdir -p /run/lock/lmlib && chown root:lock-manager /run/lock/lmlib && chmod -R 0770 /run/lock/lmlib

    2. After reboot Plesk became accessible and to make SSH work you have created the following scheduled task at Plesk Scheduled Task manager:
    mkdir /var/run/sshd && chmod 0755 /var/run/sshd && /etc/init.d/ssh start

    After that, the issue was resolved.

    This article will be updated to give more details.

  • Avatar
    Nico Weis (Edited )

    Got the same problem, PLESK noch reachable, no access by ssh

     

    Check that /repair/run/lock/lmlib and /repair/var/lock/lmlib exists 

    -> Both directorys are not existing
    Just /repair/run/lock/pmlib and /repair/var/lock/pmlib

    Any Ideas? No solution fixed my problem :( 

     

    root@h2756151:~# stat /repair/run/lock/pmlib

      File: '/repair/run/lock/pmlib'

      Size: 4096      Blocks: 8          IO Block: 4096   directory

    Device: ddf0b6e1h/3723540193d Inode: 23593373    Links: 2

    Access: (0755/drwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root)

    Access: 2018-11-27 19:36:37.696311335 +0100

    Modify: 2018-11-27 19:19:30.726094850 +0100

    Change: 2018-11-27 19:19:30.726094850 +0100

    Birth: -

  • Avatar
    Alexandr Redikultsev

    Hi @Nico Weis and @Sean Unett,

    Thank you for your feedback!

    My colleague Ivan Postnikov suggested a little bit different solution for the problem. We also sent a notice to Strato regarding this issue.

    I have also added this information to the article under "In case SSH connection to the server is not possible" section.

    Please, check it and let me know whether or not it helps.

  • Avatar
    Alexandr Redikultsev

    Hi @ Claus Siebeneicher,

    'Enhanced security mode' is related to the internal Plesk logic of storing the passwords, sending notifications and other stuff, however it does not affect other services like SSH.

    Let me know if you still face issues with the new solution suggested in the article.

  • Avatar
    Erich Schotborgh (Edited )

    Hi, this solution didn't work for me (also on a Strato VPS). However, we did finally fix it thanks to my friend Aniel:

    • Boot RESCUE SYSTEM for SSH access
    • # chroot /repair
    • Paste this in /etc/rc.local
    #!/bin/sh -e # # rc.local # # This script is executed at the end of each multiuser runlevel. # Make sure that the script will "exit 0" on success or any other # value on error. # # In order to enable or disable this script just change the execution # bits. # # By default this script does nothing. # fixes the "Missing privilege separation directory: /var/run/sshd "problem with "openssh-server amd64 1:7.2p2-4ubuntu2.6" if [ ! -d /var/run/sshd ]; then mkdir /var/run/sshd chmod 0755 /var/run/sshd systemctl enable ssh fi exit 0

    Next reboot server in normal mode. SSH should work.

    Next I executed this line to repair Plesk:

    # plesk installer --select-release-current --reinstall-patch --upgrade-installed-components

    He also made a little video to help others:

    https://www.youtube.com/watch?v=61k0UPzsTOA

    Good luck guys!

     

     

  • Avatar
    Alexandr Redikultsev

    Hi @Erich Schotborgh,

    Thanks a bunch for sharing!

  • Avatar
    Stefan Braun

    H, I am using Strato V-Server Ubuntu 16.04.5 LTS Plesk Onyx Version 17.5.3 and I have the same problem and I have no SSH connection. I have tried all recommendations to this problem in several threats without any success, especially the main recommendation in this threat (In case SSH connection to the server is not possible) but also the recommendations of @Ivan Postinok and @Erich Schotborgh.

    Even with the solution of Erich Schotborogh I cannot get a SSH connection (Error message of putty: Network error: Connection refused. [and no, it's not a firewall problem, I can get connection in the same way to my other Strato servers). Strato has no alternative way and only use the same scripts of Erich Schotborogh here (but without the line "systemctl enable ssh", also this does not work). There is no access to the server and also no access to the mail accounts (webmail / horde, error message for the mail problem is "wrong user or password", maybe this is the follower problem of missing port access).

    For any alternative solutions I really would be very thankful.

     
     
  • Avatar
    Ivan Postnikov

    Hello @Stefan, 

    Do you have an access to Plesk interface after following the instructions from the article and @Erich Schotborgh?

    According to the message "Network error: Connection refused", most probably, the issue now is with SSHD.

    If you now have an access to Plesk, the SSH may be repaired the following way:

    1. Go to Tools & Settings > Scheduled Tasks.

    2. On behalf of root run the below task:
    mkdir /var/run/sshd && chmod 0755 /var/run/sshd && /etc/init.d/ssh start

  • Avatar
    Stefan Hamann

    have already tried all the possibilities dur. Plesk program writer has no further solution ??

  • Avatar
    Pavel Rozental

    Hello @Stefan Hamann,

    If you applied steps from this article and they did not help probably you faced another issue.

    I recommend you to contact Plesk support according to the article https://support.plesk.com/hc/en-us/articles/213608509

  • Avatar
    Artyom Baranov

    @Stefan Hamann,

    Hello,

    If for some reason a hosting provider is not able to assist, you may always receive support directly from Plesk using a support subscription. Visit this KB article to learn more about this opportunity.

  • Avatar
    Stefan Hamann
    die Rechte von / waren falsch:
    chmod 777 und chown magicspam.psaadm
    
    Nach einem chmod 755 / und chown root.root / (und einem Reboot) scheint
    wieder alles zu funktionieren.

    wurde jetzt von einem externen Admin erledigt.

    die sache hat sich somit erledigt.

  • Avatar
    Artyom Baranov

    @Stefan Hamann,

    Seems that the cause of the issue you have faced and the issue described in the article are different.

    Anyway, I am glad that you have managed to resolve it!

  • Avatar
    Stefan Braun (Edited )

    Thank you very much for you suggestions. I couldn't sort out the problem in these ways. Plesk doesn't give support because of the third party licence. Strato does refer to Plesk because Strato doesn't give a serious support. I had to consult a professional external programmer. He has reinstalled the SSH. Because I also had no access to the mail server also this had to be reinstalled. Tables had to be repaired to get back access. It was complicated.

    But I understand now better the problem. I have done a reinstall out of the BackupControl from Strato-Server-Login. And this "reinstall" of the complete system did generate all these problems. The databases crashed because of the reinstall. Strato support confirms not to use the reinstall functions on Linux Servers because it doesn't work as a snapshot and it isn't possible to do a recovery on a running Linux system. But there is also no manually access to the backuped data from BackupControl, there is no chance to use this backup function. At the end of the story Strato BackupControl for Linux systems is a trap: an absolutely useless function that causes more problems than to solve them. Strato did confirm about this and I am really wondering about why they are offering this useless function. I am moving now to a provider with snapshot function and a working support in background.

    With my knowledge from today the workaround with [LockManagerException] problem could be solved in 15 minutes. But it isn't possible if you have first tried to do a reinstall out of BackupControl. Good luck and thanks.

  • Avatar
    William de Weerd

    I have absolutely no idea what to do, have already tried the above I am here in recovery what should I fill in now? already thanks for the help (Strato Server) I am a noob

    Plesk\Lock\Exception

    Lock Manager error: '[LockManagerException] Can't open or create shared memory by shm.name: "/run/lock/lmlib/SharedLockManagerStorage0.2.4"; shm.start_size: "8388608"; error "No such file or directory"'.

    I'm here now

  • Avatar
    Anton Maslov

    @William de Weerd you need to proceed exactly as per article:

    Open the below file with "vi" editor:

    vi /repair/var/spool/cron/crontabs/root

    Put there at the end the below content:

    @reboot mkdir -p /run/lock/lmlib ; chown root:lock-manager /run/lock/lmlib ; chmod -R 0770 /run/lock/lmlib ; mkdir /var/run/sshd ; chmod 0755 /var/run/sshd ; /etc/init.d/ssh restart

    Save and now reboot the server to the normal mode, that should fix the issue.

    Additionally, this is a bug in OpenVZ, problem not related to Plesk software, thus Strato should provide you with support for that.

  • Avatar
    William de Weerd

    @Anton Maslov 

  • Avatar
    Roland Finke (Edited )

    I tried the solution "in case SSH connection to the server is not possible" at the top. I could connect to the server and I could also edit the file.

    I noticed I had a typo in this: 

    @reboot mkdir -p /run/lock/lmlib ; chown root:lock-manager /run/lock/lmlib ; chmod -R 0770 /run/lock/lmlib ; mkdir /var/run/sshd ; chmod 0755 /var/run/sshd ; /etc/init.d/ssh restart

    I used chmod 0775 instead of chmod 0755

    Fixed that and ran it again.

    Still not successful!

    It had absolutely no effect! My hoster is Strato. What should I do now? Please help! 

    P.S.: My website is still available for my users but I still can't access Plesk.

    I also noticed that my error message is slightly different from the one a above. Mine is like this:

    Plesk\Lock\Exception

    Lock Manager error: '[LockManagerException] Can't open or create shared memory by shm.name: "/run/lock/lmlib/SharedLockManagerStorage0.2.4"; shm.start_size: "8388608"; error "No such file or directory"'.

    So it is not "access denied" but "no such file or directory".

    If required I could also use one of Strato's back ups and recover the server completely. But will this fix the issue? And would what would I need to do after recovering the server to avoid the problem from happening again? I have absolutely no idea of Linux commands etc. So I would need a very detailed instruction. Sorry.

     

    Please help.

  • Avatar
    Alisa Kasyanova

    @Roland Finke
    Hi! Try to do the following:

    1) Execute this command:
    sysv-rc-conf --list psa

    Check the output. It should have ON for 2, 3, 4 and 5:
    psa 0:off 1:off 2:on 3:on 4:on 5:on 6:off

    If it is not, run this command:
    sysv-rc-conf psa on

    2) Reboot the server. The needed file may have not been created properly, reboot usually helps.

    Hope the steps are detailed enough and that they will help :)

  • Avatar
    Roland Finke (Edited )

    @Alisa Kasyanova

    Thanks for this! Unfortunately it didn't work. Since I cannot connect to the server using ssh I turned on Strato's repair mode and connected to the server via PuTTY. I entered the first command. The message is "command not found". I made sure to be in the root directory using cd / tried again: not found. Then I got into the repair folder tried again: not found. Then I tried the second command you mentioned: same result!

    I am getting a little desperate with this to be honest. I am using Plesk in order to not to have to deal with cryptic Linux commands ... And even using those crypctic commands I can't get it fixed ... This is a nightmare for me.

    Is there no solution that is less complicated? e.g. you deliver a patch and with the next automatic system update it is being implemented? Something like this would be greatly appreciated! I am regularly getting notification emails from the Package Update Manager informing me about the latest system updates which I set to be implemented automatically. Can't you use the Package Update Manager to implement the required fixes? (In that case I would only have to wait one day and everything would be ok!)

    Please help! What should I do next?

    P.S.: Just noticed that the situation has become even worse: Earlier I would still be able to access the Plesk login page. No longer possilbe. I get directed to the error page immediately. In addition I was able to access PHPMyAdmin. No longer possible! Same error page. I cannot run my database back ups any longer. That really is a complete disaster for me.

     

     

  • Avatar
    Lev Iurev

    @Roland Finke

    As I can see you've created a ticket, let's continue there

  • Avatar
    Roland Finke

    Problem is solved! After buying the Plesk support I got help immediately!! I allowed the support team to access my server via rescue mode which I had enabled for them. They fixed it in no time. I rebooted in normal mode. Done.

    I would have NEVER been able to do it myself! That €10 are well invested! Will keep the subscription.

    My advice to all Strato Linux Virtual Server customers: Get the Plesk support and get in English (not German!). Then you have 365x24x7 support which is really essential if you don't want server down time during office hours!!

    Good luck to everyone facing this issue!

    Roland

  • Avatar
    Tristan Huitenga

    Seriously, buying Plesk support for this? The answer is already given. Thank you Anton Maslov. Your description was most short and worked. I tested removing the line after it all worked from the cron job. After rebooting same issue. Seems the line should stay.

    Ok for the people, like me that get anoyed when only referencing:

     

    1. Logged into Strato.

    2. Went to Server Management.

    3. Put Server in Recovery Mode.

    4. SSH't to it with # ssh root@stratohostname.stratoserver.net

    5. Removed all spaces from the password that is provided on the Strato Recovery Page you get to see when you start you're strato vps in Recovery mode.

    6. Finally connected to Repair mode with ssh:

    # vi /repair/var/spool/cron/crontabs/root

    "type i for insert, then put the following line:"

    @reboot mkdir -p /run/lock/lmlib ; chown root:lock-manager /run/lock/lmlib ; chmod -R 0770 /run/lock/lmlib ; mkdir /var/run/sshd ; chmod 0755 /var/run/sshd ; /etc/init.d/ssh restart

    "now press escape and after that :wq and ENTER, for saving."

    7. Go back to the Strato page click on the tab "RecoveryManager" again you pressed to go into recovery. A button will appear to start the server normally again.

    8. Leave the line where it is. Do not remove it afterwards. Because then you will have to do it all over again. Trust me I've been there.

  • Avatar
    Nikita Nikushkin

    Hi @Tristan Huitenga!

    Thank you for additional information regarding Strato itself

    I suppose others will find such information useful

Please sign in to leave a comment.

Have more questions? Submit a request