The portable Linux client doesn’t have C+±exception support (yet). In my tests everything was functional but this is something that must be resolved before releasing a non-beta version.
Changes with server 2.0.21 beta
Prevent client list deletion during backup
Run incremental image backup also if full image backup is disabled
Halved (meta-data) pipe file read timeout
Nightly cleanup of SYSVOL/ESP images associated with incomplete images
Use correct identity in channel thread
Only hard link files with at least 2K size
Increase max number of file index readers from 127 to 4094
Reconnect and retry if image backup block number is corrupted
Update output file size only when writing out patch and not earlier
Do not reconnect from file queuing because this causes double frees (introduced in last version)
Force client re-authentication on address change (e.g. from Internet to local)
Changes with server 2.0.20 beta
Fix: Use tar file name instead of last file name in log output
Reconnect if file request fails and retry
Changes with server 2.0.19 beta
Fix saving meta-data for files in meta-data
Changed log level of “Saved metadata of…” to debug
Fix issues with file entry index
Reintroduced the missing syncing before full database checkpoint
Prevent shared mutex writer starvation
Prevent recursive read locking of shared mutex
Resize hash output to correct size if it is lower than what was already output
Use tar file name instead of last file name in log output
Log file download corruptions
Changes with server 2.0.18 beta
Do not re-created script and meta-data transfers when client is restarted and server reconnects
Treat SQLITE_PROTOCOL error like SQLITE_BUSY (retry query)
Removed systemd task limit in systemd unit file
New file backup status “completed with issues” and OS and client version persistence
Changes with server 2.0.17 beta
Do not verify client side hashes on incomplete files
Manually resolve relative vhd paths to work around Windows path limit
Changes with server 2.0.16 beta
Handle shadow copy failures with 1.4.x client
Improved handling of case where snapshot removal is delayed because of still ongoing meta-data transfers
Improved logging when stopping meta-data transfer
Improved file meta-data shutdown procedure
Changes with server 2.0.15 beta
Fix setting non-admin password from command line
Reset exponential backoff also if backup failed with e.g. timeout
Fixed file index racing condition
Load whole block regardless of block hash if reading for hash fails
Fix os_set_file_time for FreeBSD directories
Wait for at least 10s before hard shutdown of meta-data transfer
Log about download with sparse extents
Warning instead of assert if file size is wrong in meta-data
Retry getting VSS log data
Re-authenticate with client after client update
Symbolic link creation check on status screen
Changes with client 2.0.19 beta
Lowered idle timeout for file upload from 60h to 1h
Free file after 1 min instead of 1 hour if idle
Changes with client 2.0.18 beta
Do not re-created script and meta-data transfers when client is restarted and server reconnects
Correctly cleanup file meta-data stream if interrupted and not reconnected
Changes with client 2.0.17 beta
Fix share referencing error after timeout
Revert: Fix share referencing error after reconnection
Changes with client 2.0.16 beta
Fix share referencing error after reconnection causing hang in snapshot removal
Changes with client 2.0.15 beta
Handle case where volume size is not block size (4k) aligned
Changes with client 2.0.14 beta
Fix SYSCONFDIR variable replacement in Linux scripts
Improved handling of case where snapshot removal is delayed because of still ongoing meta-data transfers
Changes with client 2.0.13 beta
Fix line endings in postgres backup scripts
List VSS provider after VSS provider error
Do not delete excluded/not included files/directories when restoring
Do not try to create windows volume as directory during restore
Correctly open alternative file for renaming later during restore
Various file restore fixes
Restore all files when restore is done via client regardless of user access tokens
Perform incremental indexes after unfinished full indexes
Todo
UEFI/GPT testing
Compatibility with prior versions
2.x server with 1.4.x client full compatibility (please report issues)
2.x client with 1.4.x server works only in local network mode (not via internet mode)
Older client/server combinations may work but were not tested
1.x restore does not work with 2.x servers (improved login method)
Upgrade process
As always: Replace the executables (via the installers) and the database of the server/client will be updated on first running it. As always downgrading the database version after upgrading it is not possible, so you should backup the old database files especially since this is a beta.
Because of the improved file de-duplication and statistics calculation the largest server table has to be completely rebuild. This may take a few hours depending on how many file entries you have. It will show the progress on the web interface but is not usable during the upgrade process.
Linux notes:
The wrapper scripts start_urbackup_server and start_urbackup_client have been removed. Please use the executable directly
The executable has been renamed to urbackupsrv (from urbackup_srv), the client to urbackupclientbackend (from urbackup_client)
There is a new command line interface for the client urbackupclientctl
All the plugins are now statically linked into one executable. This simplifies the compilation, debugging and packaging on Linux
Run the UrBackup server on Linux with e.g. urbackupsrv run --loglevel debug
Configuration: Server 2.0.16 [Debian 8.3] Problem: Server stopping, cannot open admin web page
I’ve had the server stop with no error message, but maybe this will help:
/var/log/urbackup.log:
2016-05-09 05:48:29: WARNING: Shutting down (Signal 15)
[Upgrade to 2.0.16 here]
2016-05-09 05:50:01: ERROR: Error downloading server version information: HTTP response code said error(ec=22), The requested URL returned error: 404 Not Found
2016-05-09 05:50:54: ERROR: Error in update: verify_sig_err
2016-05-09 05:50:54: ERROR: Timeout: In client update
2016-05-09 05:58:49: WARNING: Restarting shadow copy of D:\ because it was started by this server
2016-05-09 06:01:19: WARNING: Restarting shadow copy of D:\ because it was started by this server
2016-05-09 06:50:03: ERROR: Error downloading server version information: HTTP response code said error(ec=22), The requested URL returned error: 404 Not Found
2016-05-09 06:50:05: ERROR: No permission to access “/data/urbackup/CLIENT/160508-0352”
2016-05-09 06:50:05: WARNING: Warning: Directory doesn’t exist: “/data/urbackup/CLIENT/160508-0352”
^C
systemctl status urbackupsrv
urbackupsrv.service - LSB: Server for doing backups
Loaded: loaded (/etc/init.d/urbackupsrv)
Active: active (exited) since Mon 2016-05-09 05:48:59 PDT; 1h 9min ago
When this happens, I have to run systemctl restart urbackupsrv to get working again
You could put it into debug log mode (in /etc/default/urbackupsrv) and/or run it in a debugger after compiling it (run with gdb --args urbackupsrv run -v debug – after crash enter bt), or configure it such that it creates core dumps and send me those.
Okay so the Stop button doesn’t work but restarting the “UrBackup Client Service for Backups” on the Windows client did the trick. Backup succeed but with error :
[ 242.116567] BTRFS info (device sdb): The free space cache file (559985721344) is invalid. skip it
[ 242.189118] BTRFS info (device sdb): The free space cache file (561059463168) is invalid. skip it
[ 242.307136] BTRFS info (device sdb): The free space cache file (562133204992) is invalid. skip it
[ 657.659983] fileindex write[776]: segfault at 0 ip 000000000066f051 sp 00007efdeefb1430 error 4 in urbackupsrv[400000+639000]
2016-05-11 13:14:41: WARNING: File entry with id 4629523 with filesize 28672 found in entry index while deleting, but should be there. The file entry index may be damaged.
2016-05-11 13:14:41: WARNING: File entry with id 4632831 with filesize 1572864 found in entry index while deleting, but should be there. The file entry index may be damaged.
2016-05-11 13:14:41: WARNING: File entry with id 4632832 with filesize 32768 found in entry index while deleting, but should be there. The file entry index may be damaged.
2016-05-11 13:15:00: ERROR: LMDB: Failed to delete data (MDB_NOTFOUND: No matching key/data pair found)
2016-05-11 13:15:00: ERROR: LMDB: Failed to delete data (MDB_NOTFOUND: No matching key/data pair found)
2016-05-11 13:15:00: ERROR: LMDB: Failed to delete data (MDB_NOTFOUND: No matching key/data pair found)
2016-05-11 13:15:08: ERROR: LMDB had error during increase (on commit). Aborting…
Update:
I also see this in the activity log just before the server crashes:
05/11/16 16:17 INFO Waiting for file hashing and copying threads…
05/11/16 16:17 INFO Waiting for metadata download stream to finish
05/11/16 16:17 INFO Writing new file list…
05/11/16 16:18 DEBUG Some metadata was missing
05/11/16 16:18 INFO Number of copyied file entries from last backup is 3
05/11/16 16:18 DEBUG Client disconnected while backing up. Copying partial file…
05/11/16 16:18 INFO Transferred 33.6981 MB - Average speed: 200.216 KBit/s
05/11/16 16:18 INFO (Before compression: 96.5814 MB ratio: 2.86608)
05/11/16 16:18 INFO 13.3232 GB of files were already present on the server and did not need to be transferred
05/11/16 16:18 DEBUG Script does not exist urbackup/post_incr_filebackup
05/11/16 16:18 INFO Time taken for backing up client CLIENT: 45m 57s
05/11/16 16:18 ERROR Backup failed
getting this error since update to latest versions…
madiadb dump is enabled in /usr/local/etc/urbackup/madiadbdump.conf
initial backup was fine. but incremental throws this
Errors 12.05.16 11:12 Writing metadata to /mnt/Backups/jumpfwdb.hotline.local/160512-1112/.hashes/urbackup_backup_scripts/mariadbdump.sql failed
Errors 12.05.16 11:13 Fatal error during backup. Backup not completed
Errors 12.05.16 11:13 FATAL: Backup failed because of disk problems
Errors 12.05.16 11:13 Backup failed
I can guarantee that the Harddisks are absolutly fine…
I use the prefilebackup.bat and postfilebackup.bat on clients to stop and start/run scripts before and after backups since around 2.0.16beta client the postfilebackup.bat is not running after the backup completes however the prefilebackup.bat runs.
