Standalone backup agents for Linux, Mac, AIX & Solaris workloads on-premises or in the public cloud
Locked
PTide
Product Manager
Posts: 6408
Liked: 724 times
Joined: May 19, 2015 1:46 pm
Contact:

[LINUX] Known Issues

Post by PTide » 2 people like this post

General
•User interface may appear unreadable in some terminal programs due to using incorrect symbols for pseudographics. To fix the issue, do the following:

Putty: execute

Code: Select all

echo export "NCURSES_NO_UTF8_ACS=1" >> ~/.bashrc
Kitty: select the Allow ACS line drawing in UTF check box.

Backup

• Volume-level backup job relies on a device name under the /dev directory and requires volumes to have their corresponding device names under the /dev directory staying persistent (e.g. /dev/md-127, /dev/dm-1). Otherwise the job will back up the wrong volume.
• Backup to CIFS is not supported for x86+PAE systems.
• Overall amount of file-level backup data is limited to 4.0 TB, including filesystem structures (such as inodes and blocks of pointers).
• Cloud Connect certificates are not verified neither against CRL, nor through OCSP.
• Files residing on the directories encrypted with eCryptfs cannot be selected for file-level backup.
• Data encrypted with eCryptfs stays encrypted inside a volume-level backup.
• Network shares (both CIFS and NFS) can be selected for backup only in snapshot-less mode.
• RAID members cannot be backed up.
• Disks partitioned with cfdisk version older than 2.25 are not supported.
• File level backup masks do not apply to directories.
• Backup job type cannot be switched between file and volume levels.
• Very high data change rate during backup may cause snapshot overflow. Should this happen, the job will attempt to retry processing up to 3 times with a larger snapshot.

Restore
• File-level restore cannot be performed on kernels versions 4.0.x-4.1.33 due to the bug in FUSE kernel module. To work around the issue, either upgrade the kernel, or perform file-level restore from the Recovery Media.
• File-level restore for files residing on eCryptfs directories does not mount eCryptfs filesystem automatically.
• Concurrent file-level restore sessions from different restore points are not supported because only one restore point can be mounted at a time. To perform a file-level restore from another restore point, unmount currently mounted one first.
• Restore of encrypted devices is not supported.

Veeam Backup & Replication Integration
• Encrypted backups produced with v3.0 can be imported to Veeam Agent for Linux directly.
• Fast clone is not supported for ReFS repositories.

Oracle AAIP
Databases residing on ASM volumes cannot be backed up. However, they will be prepared by AAIP engine and will be shown in Veeam Explorer for Oracle without possibility to be restored.

Configurations
For GPT disks: bootloader will not be backed up if BTRFS root and BIOS /boot reside on different physical disks partitions.
BTRFS Support
• Bugs and issues in the kernel that affect BTRFS snapshots and send/receive functionality can cause backup jobs to fail.
• Backup progress for subvolumes cannot be properly shown on systems with btrfsprogs versions lower than v4.1.
• Large overhead can be observed in the Transferred statistics field for backup jobs that include BTRFS that has no subvolumes.
• If the root subvolume (id 5) is more than 50% full, restore of the subvolume may fail. For for details and workaround, see the following Veeam Knowledge Base article: https://www.veeam.com/kb2877.
• If a subvolume is deleted during restore, it is required to remove the corresponding entry from fstab. Otherwise, the system will not boot.

Snapshot-Less Mode
• Indexed BTRFS mount points cannot be browsed or restored in Veeam Backup Enterprise Manager.
• The tmpfs file system cannot be backed up in the snapshot-less mode.

ISO
UI may not show backups that contain localized symbols in their names.

Agent Management
• After upgrade to U4, for Agent Management jobs that contain Linux servers running Oracle databases it is required to either specify valid Oracle credentials in the job’s Application-Aware Processing settings, or to disable Oracle processing. Otherwise the job will fail.
• Existing backup policies will stop working and receive updates if the backup server IP address changes.
• Statistics for file-level backup jobs managed by the backup server may be inaccurate.
• A backup job managed by the backup server will perform merge even if a corrupted restore point is present in a backup chain. The corrupted restore point will be deleted during the next job run.
Other
• A backup job that uses a network folder as a target cannot be created on systems with Kaspersky Endpoint Security 10.
• Installation of the veeamsnap kmp package causes kernel 4.12.14-lp150.11-default to be installed on openSUSE Leap 15.0.
• If an LVM physical volume is selected for backup, all volume groups that fully reside in this physical volume will be backed up.

Installation
For dkms version of veeamsnap module only:
• dkms package is not present in default repositories for some Linux distributions. You should obtain it from third-party repositories:
CentOS / RHEL / Oracle Linux / Fedora : EPEL repository
OpenSUSE / SLES : Packman repository

For openSUSE/SLES: dkms from the Packman repository contains wrong dependencies so if you see

Code: Select all

Problem: nothing provides kernel-devel needed by dkms-2.2.0.3-14.1.noarch
Solution 1: do not install dkms-2.2.0.3-14.1.noarch
Solution 2: break dkms-2.2.0.3-14.1.noarch by ignoring some of its dependencies
then make sure that you have an appropriate kernel-devel package installed and its version matches your kernel version:

Code: Select all

root@localhost:~> rpm -qa | grep kernel-default
kernel-default-devel-3.0.101-91.1
kernel-default-3.0.101-91.1
and tell zypper to ignore dependencies:

Code: Select all

zypper -n install --force dkms
• kernel-devel version must match your current kernel version.
Extended kernels such as kernel-pae, kernel-uek, and others, require appropriate kernel-devel packages to be installed, for example:

kernel-uek > kernel-uek-devel, kernel-default > kernel-default-devel, kernel-pae > kernel-pae-devel.
To check your current kernel version, execute

Code: Select all

uname -r
• RHEL and derivatives: if yum package manager installs packages that do not match your current kernel version, either update your system or fetch older versions of the required packages from CentOS Vault repository.

• SLES/OES: Veeam Agent for Linux was successfully installed but you see the error

Code: Select all

|--tr:Failed to execute [modprobe].
[05.01.2017 08:04:48] <140669894625024> lpbcore| >>  |Failed to load module [veeamsnap] with parameters [zerosnapdata=1
To fix the issue edit /etc/modprobe.d/unsupported-modules.conf and change the value of the variable to 1
PTide
Product Manager
Posts: 6408
Liked: 724 times
Joined: May 19, 2015 1:46 pm
Contact:

New Issues

Post by PTide »

If you use Veeam Agent for Linux with a paid license installed, you can submit a support case to the Veeam Support Team at www.veeam.com/support.html. If you use Veeam Agent for Linux FREE, please submit issues by creating a separate topic for each issue. Please do not post log snippets, instead include the link to full logs package.

To export logs:

1. Launch the Veeam Agent for Linux control panel with the veeam or veeamconfig ui command.

2. In the Veeam Agent for Linux control panel, press the 'm' key to open the Miscellaneous menu.

3. In the menu, select the Export Logs option and press Enter.

4. In the Choose logs directory window, browse to a directory to which you want to export logs and press Ok to start logs export process.

Veeam Agent for Linux will collect product logs, export them to an archive file with the name veeam_logs.tar.gz, and save the archive file to the specified directory.
Locked

Who is online

Users browsing this forum: No registered users and 15 guests