Skip to main content

New CPanel install: Quotas impossible to fix

Mise
I have opened a ticket support. Please, delete this post

Comments

33 comments

Date Votes
  • cPRex Jurassic Moderator
    Hey there! Rather than removing the post you always have the option to keep the original errors you're seeing posted, and also post the ticket number (if it was opened with cPanel) so we can track this on our end. That way we can post the resolution and it may help someone else out in the future.
    0
  • Mise
    Ticket ID was 94304948. Although I'm sorry to say it was not very effective at those moments, with some pressure on my side to build a new server quickly because the OVH fire. And I leaved the ticket. Now is closed The new server is working although with quotas still in error. These are the logs and commands with relevant messages: quotas not working: # quotacheck -avgum quotacheck: Cannot find filesystem to check or filesystem not mounted with quota option.
    dmesg shows Disk quotas dquot_6.5.2 is present: # /var/log/dmesg [ 0.655995] HugeTLB registered 2 MB page size, pre-allocated 0 pages [ 0.656974] zpool: loaded [ 0.656978] zbud: loaded [ 0.657180] VFS: Disk quotas dquot_6.5.2 [ 0.657203] Dquot-cache hash table entries: 512 (order 0, 4096 bytes) [ 0.657366] Key type big_key registered [ 0.657369] SELinux: Registering netfilter hooks [ 0.658406] NET: Registered protocol family 38
    quotas are configured: # /var/log/quota_enable.log journaled quota support: kernel supports, user space tools supports (available) UUID=6cd50e51-cfc6-40b9-9ec5-f32fa2e4ff02 (enabling quotas) The system will configure quotas on the "UUID=6cd50e51-cfc6-40b9-9ec5-f32fa2e4ff02" which is using the "xfs" filesystem. A reboot will be required to enable quotas on xfs. Updating Quota Files..........Done Quotas have been enabled and updated. Modifying the /etc/default/grub file to enable user quotas... Running the "grub2-mkconfig" command to regenerate the system's boot configuration... Generating grub configuration file ... Found linux image: /boot/vmlinuz-3.10.0-1127.19.1.el7.x86_64 Found initrd image: /boot/initramfs-3.10.0-1127.19.1.el7.x86_64.img Found linux image: /boot/vmlinuz-3.10.0-1127.el7.x86_64 Found initrd image: /boot/initramfs-3.10.0-1127.el7.x86_64.img Found linux image: /boot/vmlinuz-0-rescue-cab9605edaa5484da7c2f02b8fd10762 Found initrd image: /boot/initramfs-0-rescue-cab9605edaa5484da7c2f02b8fd10762.img done The '/' partition uses the XFS. filesystem. You must reboot the server to enable quotas.
    audit and grub: # /var/log/audit/audit.log :type=SERVICE_START msg=audit(1615572032.527:97): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=cpanelquotaonboot comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success' # /var/log/grubby DBG: linuxefi /vmlinuz-3.10.0-1127.19.1.el7.x86_64 root=UUID=0ef2a656-b53f-49a2-8478-9a301b3b0617 ro rd.auto crashkernel=auto vga=normal nomodeset rootflags=uquota
    cloud-init.log # /var/log/cloud-init.log 2021-03-12 18:00:07,390 - util.py[DEBUG]: Fetched {'configfs': {'mountpoint': '/sys/kernel/config', 'opts': 'rw,relatime', 'fstype': 'configfs'}, 'efivarfs': {'mountpoint': '/sys/firmware/efi/efivars', 'opts': 'rw,nosuid,nodev,noexec,relatime', 'fstype': 'efivarfs'}, '/dev/loop0': {'mountpoint': '/var/tmp', 'opts': 'rw,nosuid,noexec,relatime,discard,data=ordered', 'fstype': 'ext4'}, 'devpts': {'mountpoint': '/dev/pts', 'opts': 'rw,nosuid,noexec,relatime,gid=5,mode=620,ptmxmode=000', 'fstype': 'devpts'}, 'debugfs': {'mountpoint': '/sys/kernel/debug', 'opts': 'rw,relatime', 'fstype': 'debugfs'}, 'securityfs': {'mountpoint': '/sys/kernel/security', 'opts': 'rw,nosuid,nodev,noexec,relatime', 'fstype': 'securityfs'}, 'sysfs': {'mountpoint': '/sys', 'opts': 'rw,nosuid,nodev,noexec,relatime', 'fstype': 'sysfs'}, 'mqueue': {'mountpoint': '/dev/mqueue', 'opts': 'rw,relatime', 'fstype': 'mqueue'}, 'pstore': {'mountpoint': '/sys/fs/pstore', 'opts': 'rw,nosuid,nodev,noexec,relatime', 'fstype': 'pstore'}, '/dev/sda1': {'mountpoint': '/boot/efi', 'opts': 'rw,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=ascii,shortname=mixed,errors=remount-ro', 'fstype': 'vfat'}, 'hugetlbfs': {'mountpoint': '/dev/hugepages', 'opts': 'rw,relatime', 'fstype': 'hugetlbfs'}, 'systemd-1': {'mountpoint': '/proc/sys/fs/binfmt_misc', 'opts': 'rw,relatime,fd=31,pgrp=1,timeout=0,minproto=5,maxproto=5,direct,pipe_ino=13075', 'fstype': 'autofs'}, 'cgroup': {'mountpoint': '/sys/fs/cgroup/pids', 'opts': 'rw,nosuid,nodev,noexec,relatime,pids', 'fstype': 'cgroup'}, 'sunrpc': {'mountpoint': '/var/lib/nfs/rpc_pipefs', 'opts': 'rw,relatime', 'fstype': 'rpc_pipefs'}, '/dev/md2': {'mountpoint': '/boot', 'opts': 'rw,relatime,attr2,inode64,noquota', 'fstype': 'xfs'}, 'tmpfs': {'mountpoint': '/sys/fs/cgroup', 'opts': 'ro,nosuid,nodev,noexec,mode=755', 'fstype': 'tmpfs'}, 'proc': {'mountpoint': '/proc', 'opts': 'rw,nosuid,nodev,noexec,relatime', 'fstype': 'proc'}, 'devtmpfs': {'mountpoint': '/dev', 'opts': 'rw,nosuid,size=16217048k,nr_inodes=4054262,mode=755', 'fstype': 'devtmpfs'}, '/dev/md3': {'mountpoint': '/', 'opts': 'rw,relatime,attr2,inode64,noquota', 'fstype': 'xfs'}, 'rootfs': {'mountpoint': '/', 'opts': 'rw', 'fstype': 'rootfs'}} mounts from proc
    boot messages: # /var/log/messages Mar 12 18:59:54 host kernel: VFS: Disk quotas dquot_6.5.2 Mar 12 19:00:16 host systemd: Starting cPanel fix quotas on boot... Mar 12 19:00:19 host dracut: Executing: /usr/sbin/dracut --quiet --hostonly --hostonly-cmdline --hostonly-i18n --hostonly-mode strict -o "plymouth dash resume ifcfg" --mount "/dev/disk/by-uuid/0ef2a656-b53f-49a2-8478-9a301b3b0617 /sysroot xfs defaults,uquota" --no-hostonly-default-device -f /boot/initramfs-3.10.0-1127.19.1.el7.x86_64kdump.img 3.10.0-1127.19.1.el7.x86_64 /var/log/messages:Mar 12 19:00:32 host fixquotas-onboot: You must reboot the server to enable XFS. filesystem quotas. /var/log/messages:Mar 12 19:00:32 host systemd: Started cPanel fix quotas on boot. ... 2021-03-12 18:00:15,683 - cc_growpart.py[DEBUG]: '/' SKIPPED: device_part_info(/dev/md3) failed: /dev/md3 not a partition 2021-03-12 18:00:15,698 - main.py[DEBUG]: Ran 12 modules with 0 failures 2021-03-12 18:00:17,111 - main.py[DEBUG]: Ran 13 modules with 0 failures 2021-03-12 18:00:19,061 - main.py[DEBUG]: Ran 10 modules with 0 failures
    kernel support quotas: # /usr/local/cpanel/scripts/fixquotas journaled quota support: kernel supports, user space tools supports (available) UUID=0ef2a656-b53f-49a2-8478-9a301b3b0617 (already configured quotas = 1). UUID=6fa0c2e6-8018-4134-bc19-04ede1f33c05 (already configured quotas = 0). Updating Quota Files..........Done Quotas have been enabled and updated. You must reboot the server to enable XFS" filesystem quotas.
    fstab and xfs: # cat /etc/fstab UUID=0ef2a656-b53f-49a2-8478-9a301b3b0617 / xfs defaults,uquota 0 1 UUID=6fa0c2e6-8018-4134-bc19-04ede1f33c05 /boot xfs defaults 0 0 LABEL=EFI_SYSPART /boot/efi vfat defaults 0 1 UUID=eed9d5c5-223b-45ac-a8fb-748168e79183 swap swap defaults 0 0 UUID=c3b07c38-c633-4e79-850a-5ebd2608c166 swap swap defaults 0 0 /usr/tmpDSK /tmp ext3 defaults,noauto 0 0
    quota_enable.log # cat /var/log/quota_enable.log journaled quota support: kernel supports, user space tools supports (available) UUID=6cd50e51-cfc6-40b9-9ec5-f32fa2e4ff02 (enabling quotas) The system will configure quotas on the "UUID=6cd50e51-cfc6-40b9-9ec5-f32fa2e4ff02" which is using the "xfs" filesystem. A reboot will be required to enable quotas on xfs. Updating Quota Files..........Done Quotas have been enabled and updated. Modifying the /etc/default/grub file to enable user quotas... Running the "grub2-mkconfig" command to regenerate the system's boot configuration... Generating grub configuration file ... Found linux image: /boot/vmlinuz-3.10.0-1127.19.1.el7.x86_64 Found initrd image: /boot/initramfs-3.10.0-1127.19.1.el7.x86_64.img Found linux image: /boot/vmlinuz-3.10.0-1127.el7.x86_64 Found initrd image: /boot/initramfs-3.10.0-1127.el7.x86_64.img Found linux image: /boot/vmlinuz-0-rescue-cab9605edaa5484da7c2f02b8fd10762 Found initrd image: /boot/initramfs-0-rescue-cab9605edaa5484da7c2f02b8fd10762.img done The '/' partition uses the XFS" filesystem. You must reboot the server to enable quotas.
    grub (I have tried the rebuild several times) # cat /etc/sysconfig/grub GRUB_TIMEOUT=1 GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)" GRUB_DEFAULT=saved GRUB_DISABLE_SUBMENU=true GRUB_TERMINAL="serial console" GRUB_SERIAL_COMMAND="serial" GRUB_DISABLE_RECOVERY="true" GRUB_DISABLE_LINUX_UUID="false" GRUB_CMDLINE_LINUX="rd.auto crashkernel=auto vga=normal nomodeset rootflags=uquota" GRUB_TERMINAL_OUTPUT="" GRUB_ENABLE_BLSCFG=false
    more: # ls -l /*.user ls: cannot access /*.user: No such file or directory # mount | grep noquota /dev/md3 on / type xfs (rw,relatime,attr2,inode64,noquota) /dev/md2 on /boot type xfs (rw,relatime,attr2,inode64,noquota) No quotas detected with cpanel script: # /usr/local/cpanel/scripts/resetquotas Resetting quota for userXXX to 10240 M No filesystems with quota detected. Resetting quota for computes to 1000 M No filesystems with quota detected.
    No idea really. I have found in internet some people with a similar problem with quotas using Debian. It seem their server provider failed to include the packet linux-image-extra-virtual inside their own iso's. They solved the error after installing this packet. I suspect it can be also OVH, they use its own iso's and not the first time with these problems. I wonder if perhaps there is the same problem with CentOS. The packet linux-image-extra-virtual is a superset of the kernel for Debian distros. It is not part of CentOS. However, maybe something similar can happens here. I don't know, really Any help would be appreciated.
    0
  • cPRex Jurassic Moderator
    Sorry to hear you're still having issues. Since the machine is using XFS, did you try the steps outlined here?
    0
  • Mise
    Sorry to hear you're still having issues. Since the machine is using XFS, did you try the steps outlined here?
    0
  • cPRex Jurassic Moderator
    Are the "/dev/mapper/centos_whm1-... " something that should be present?

    That's normal - it's just the name of the partition you're seeing, which will vary on each machine. There are known issues with some of the OVH kernels, so it's possible that is the root of the issue as well, but I hope they'll be able to get you better details on that.
    0
  • Mise
    ok, thanks. I will try with OVH and post here any news
    0
  • cPRex Jurassic Moderator
    That sounds good - let us know what they have to say!
    0
  • Mise
    Hi, perhaps you can help me to solve this issue. I have find in internet about Selinux can interfer in mounting quotas. I know Cpanel disble Selinux, however Inside the logs I see this sequence: # cat /var/log/demesg [ 0.000557] Security Framework initialized [ 0.000563] SELinux: Initializing. [ 0.000570] SELinux: Starting in permissive mode [ 0.000570] Yama: becoming mindful. ..... [ 2.717217] random: systemd: uninitialized urandom read (16 bytes read) [ 2.728484] systemd[1]: systemd 219 running in system mode. (+PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 -SECCOMP +BLKID +ELFUTILS +KMOD +IDN) [ 2.750214] systemd[1]: Detected architecture x86-64. [ 2.761098] systemd[1]: Running in initial RAM disk. ..... [ 4.245534] sd 6:0:0:0: [sda] Attached SCSI disk [ 4.582737] md/raid1:md3: active with 2 out of 2 mirrors [ 4.590622] md/raid1:md2: active with 2 out of 2 mirrors [ 4.598120] md3: detected capacity change from 0 to 3999041126400 [ 4.598917] md2: detected capacity change from 0 to 536281088 [ 4.629658] random: fast init done [ 4.711154] random: crng init done [ 4.987533] SGI XFS with ACLs, security attributes, no debug enabled [ 4.996805] XFS (md3): Mounting V4 Filesystem [ 5.169382] XFS (md3): Ending clean mount [ 5.715538] systemd-journald[134]: Received SIGTERM from PID 1 (systemd). [ 6.196739] SELinux: Disabled at runtime. [ 6.203327] SELinux: Unregistering netfilter hooks
    it seems Selinux is enabled at boot, then there is an attempt to mount with XFS, and later Selinux is disabled. Therefore Selinux wouldn't be disabled when the XFS mounts, impeding the quotas mounting. Is there some way to disable Selinux at starting? I don't know what process is configured to start Selinux at boot. Inside /root I can find these mentions to Selinux: # dmesg | grep -ir selinux .cpanel/datastore/_usr_bin_gtar_--help: --no-selinux Disable the SELinux context support .cpanel/datastore/_usr_bin_gtar_--help: --selinux Enable the SELinux context support tmp/yum_save_tx.2020-11-09.03-12.UDbNmt.yumtx:mbr: libselinux-devel,x86_64,0,2.5,15.el7 70 tmp/yum_save_tx.2020-11-09.03-12.UDbNmt.yumtx: relatedto: libselinux-devel,x86_64,0,2.5,15.el7@a:dependson tmp/yum_save_tx.2020-11-09.03-12.UDbNmt.yumtx: depends_on: libselinux-devel,x86_64,0,2.5,15.el7@a anaconda-ks.cfg:# SELinux configuration anaconda-ks.cfg:selinux --enforcing anaconda-ks.cfg:echo "Fixing SELinux contexts." original-ks.cfg:selinux --enforcing original-ks.cfg:echo "Fixing SELinux contexts."
    How can I locate the process starting Selinux before XFS starts? thanks
    0
  • cPRex Jurassic Moderator
    We have a guide on how to make that adjustment here:
    0
  • Mise
    yes, sestatus returns "disable". Please, see the logs the line: [ 0.000570] SELinux: Starting in permissive mode It seems Selinux starts at boot with "permissive" mode and impeding XFS, and later it is fixed like disabled, although XFS was not mounted. My question is how can I impede Selinux starting like "permissive" at boot thanks!
    0
  • cPRex Jurassic Moderator
    That part I'm not sure - if it's disabled, I would not expect that to be starting at all, but that would be something for your hosting provider to check as that isn't part of cPanel but part of the operating system itself. Can you have them look into that and see what they have to say?
    0
  • Mise
    yes, now it seems clear it is not a CPanel failure but an error in the OS installation process Still I'm waiting a definitive answer by their side... :(.
    0
  • cPRex Jurassic Moderator
    Hopefully they'll be able to get you more details on that soon!
    0
  • Mise
    two months later finally they have manifested to say I can reinstall the server because they cannot offer another solution. The non-existent support can be explained because it seems the operator chose a wrong template to install the dedicated server. According install logs, he installed several cloud services and other thing, as if the server should be devoted to a cloud network or something similar. That install template configured the boot process and selinux behavior in the very beginning, therefore forever and ever, and now Cpanel is not able to fix quotes problem. I dont' have idea how to bypass this problem still more when OVH kernels and boots are modified by themselves Also still we are waiting the promised return of money because the fire. I had a better support in AliExpress with the EVA bot to return a 0.8$ screw than in OVH spending more than 1.000$ / year thanks anyway, not the CPanel fault.
    0
  • cPRex Jurassic Moderator
    Wow - while I'm happy to hear it wasn't a cPanel issue, I'm sorry you had to go through that experience. Hopefully they make it right.
    0
  • syedc
    I realise that this is an old thread - BUT this is the thread I kept landing on, so imagine others will too. So the issue is:
    • Have a dedicated server from OVH (might also be the case with servers from KimSufi or So You Start)
    • You installed CentOS or AlmaLinux or Cloudlinux via the OVH interface
    • Now you can't get disk quota to load on Cloudlinux / cPanel
    • You've followed the cPanel guides listed here (for both EFI and UEFI systems)
    • You've engaged Cloudlinux, cPanel, OVH and your System Administrators - none of which can solve the issue bar "reformat and start again" (not an option for me)
    The fix
    • For whatever reason, OVH provisioned servers try to bootstrap the system from a networked drive (I imagine a common bootloader for their servers)
    • So all the changes you are making to your grub files simply aren't used!
    • It comes down to a file called /.ovhrc, which in my case had a "BUILD_UUID" param set to a drive I didn't recognise.
    • I edited the file to use my /boot drive's UUID
    • Regenerated my grub files with "grub2-mkconfig -o /boot/efi/EFI/almalinux/grub.cfg" (running almalinux / CL 8)
    • Had previously already edited the /etc/fstab and /etc/default/grub adding quota flags (otherwise you have to do that. Follow previously linked guides on this thread)
    • Reboot and enjoy quotas!!!!!!
    0
  • FadiObaji
    I realise that this is an old thread - BUT this is the thread I kept landing on, so imagine others will too. So the issue is:
    • Have a dedicated server from OVH (might also be the case with servers from KimSufi or So You Start)
    • You installed CentOS or AlmaLinux or Cloudlinux via the OVH interface
    • Now you can't get disk quota to load on Cloudlinux / cPanel
    • You've followed the cPanel guides listed here (for both EFI and UEFI systems)
    • You've engaged Cloudlinux, cPanel, OVH and your System Administrators - none of which can solve the issue bar "reformat and start again" (not an option for me)
    The fix
    • For whatever reason, OVH provisioned servers try to bootstrap the system from a networked drive (I imagine a common bootloader for their servers)
    • So all the changes you are making to your grub files simply aren't used!
    • It comes down to a file called /.ovhrc, which in my case had a "BUILD_UUID" param set to a drive I didn't recognise.
    • I edited the file to use my /boot drive's UUID
    • Regenerated my grub files with "grub2-mkconfig -o /boot/efi/EFI/almalinux/grub.cfg" (running almalinux / CL 8)
    • Had previously already edited the /etc/fstab and /etc/default/grub adding quota flags (otherwise you have to do that. Follow previously linked guides on this thread)
    • Reboot and enjoy quotas!!!!!!

    Hey man, I also am facing the same issue with quotas on OVHCLOUD dedicated server, I've found that file and changed UUID to my boot drive but didn't work. Can you elaborate on your solution please?
    0
  • reficul
    The fix
    • For whatever reason, OVH provisioned servers try to bootstrap the system from a networked drive (I imagine a common bootloader for their servers)
    • So all the changes you are making to your grub files simply aren't used!
    • It comes down to a file called /.ovhrc, which in my case had a "BUILD_UUID" param set to a drive I didn't recognise.
    • I edited the file to use my /boot drive's UUID
    • Regenerated my grub files with "grub2-mkconfig -o /boot/efi/EFI/almalinux/grub.cfg" (running almalinux / CL 8)
    • Had previously already edited the /etc/fstab and /etc/default/grub adding quota flags (otherwise you have to do that. Follow previously linked guides on this thread)
    • Reboot and enjoy quotas!!!!!!

    Same here, but following your instruction I not fix my quota issue. Are your .ovhrc in / ? mine is on /root folder.
    0
  • Arvy
    Not working for me too. It's a Rocky Linux 8.7 on OVH. Now (cPanel v106) AlmaLinux 9.1, Rocky 9 or Debian 20-11 are not available. OVH had only option to run under Rocky 8.7. Changing /root/.ovhrc had no effect. Grub.cfg with quotas enabled. Fstab too. Backups disabled (due to fixquotas doesn't let enable quotas in the backup partition). It's a 4x2 Tb disks server. Maybe an option is use 2 disks for / and 2 disks for /home: - install Linux, using OVH panel, only in 2 HDDs (sda and sdb). - create a new RAID using sdc and sdd:
    0
  • sukil
    Hi, Its an old thread. I am stuck with the same issue with a dedicated server from OVH and having installed their Almalinux + cPanel image. Despite doing everything quotas are not working. The fix to edit the /root/.ovhrc is not working for me too. Anybody resolved this? Thanks!
    0
  • cPRex Jurassic Moderator
    Can you let me know the output from /scripts/initquotas ?
    0
  • sukil
    Here it is: [root@s99 ~]# /scripts/initquotas journaled quota support: kernel supports, user space tools supports (available) UUID=##UUID1## (already configured quotas = 1). UUID=##UUID2## (already configured quotas = 0). Updating Quota Files..........Done Quotas have been enabled and updated.
    0
  • cPRex Jurassic Moderator
    That looks good to me - what are those UUIDs? You'd want to make sure one of them is either the entire disk or the drive where your accounts are stored.
    0
  • sukil
    That is what it is. It is a new install from the OVH almalinux + cpanel image. I followed the following article:
    0
  • cPRex Jurassic Moderator
    Are you able to edit the files mentioned in step 3 in that article?
    0
  • sukil
    Yes, and /etc/sysconfig/grub has rootflags=uquota added. [root@s28 ~]# [root@s28 ~]# cat /etc/sysconfig/grub GRUB_TIMEOUT=1 GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)" GRUB_DEFAULT=saved GRUB_DISABLE_SUBMENU=true GRUB_DISABLE_RECOVERY="true" GRUB_ENABLE_BLSCFG=true GRUB_CMDLINE_LINUX="rd.auto nomodeset console=tty0 console=ttyS1,115200n8 rootflags=uquota"
    0
  • cPRex Jurassic Moderator
    It might be best to create a support ticket with our team at this point so we can take a look.
    0
  • sukil
    I have raised a ticket 3 days back and today it got escalated to Level 2. I am waiting for a response from them.
    0
  • cPRex Jurassic Moderator
    Could you post the ticket number here so I can follow along?
    0
  • sukil
    The ticket ID is 95085007. The response from the L2 team has been great with lots of detail showing that the server has the perfect config but during boot it is using a different setup which is not clear from where. I have checked that the server is configured to boot from HDD from the OVH control panel. I have forwarded the detailed data to the OVH support team in their respective support ticket and presently I am waiting for an answer from them on the issue. One thing is really weird, the /root/.ovhrc file gets automatically updated with some odd UUID despite me updating it with the /boot UUID of my dedicated server.
    0

Please sign in to leave a comment.

Didn't find what you were looking for?

New post