I'm looking to set up a schedulled backup from command line.
I haven't found the correct command, could you tell me if it's possible and how ?
Additional question: I'm looking too, to set for the server the max number backup file to save, do you know how to do it from cli ?
Is there a command line way of removing Plesk backups?
I'm looking for the command line equivalent of these steps: Log into Plesk as the server admin > "Tools & Settings" > "Tools & Resources" > "Backup Manager" > check a backup and click "Remove".
We have a script that copies the contents of "/var/lib/psa/dumps/" to Amazon S3 for long-term storage of backups. It runs nightly after a scheduled Plesk backup. Once that script completes successfully, we'd like to delete the Plesk backup to free up disk space. We have Plesk's "Maximum number of backups in the repository" set to 1.
But Plesk appears to keep that backup around while creating a new backup (which makes some sense). This is causing us trouble because the server does not have enough disk space for two backups. As a temporary solution, I'm manually deleting the backup each day through the Plesk admin, but would like to automate it if possible. If automation is overly complex/risky, we'll look at upping our disk space instead.
I'm using this script as a cron to generate DKIM keys for the domains I have. It's an easy script. And it does it's job just OK. I have the necessary software installed and configured.
The way it works is:
The script check if the domain in queue has a record for it. If it doesn't, OpenDKIM generates a one time DKIM record for mail._domainkey that I am entering to my dns records for all the domains I have.
After creating the DKIM record, it checks and deletes mail._domainkey records if there's any. Then adds the generated DKIM to plesk dns zones and disables/enables greylisting to create the appropriate dns entry.
If it does have a DKIM recors on the hdd already, then it removes
mail._domainkeyand
TXT o=-entries.
After that it basically adds the already generated DKIM record and it disables/enables the greylisting for each domain in order to recreate "TXT o=-"
Up till here, it works fine. I can see these entries when I check the DNS Settings. They are there.
But whenever I run this script either via cron or manually, I get an error message on my panel: (You can check the screenshot as well)
Code:
Warning: The DNS zone was modified. If you would like to apply DNS template changes to this zone, either click the 'Apply DNS Template Changes' button on this page or choose the 'Apply the changes to all zones' option in Server Administration Panel > Tools & Settings > DNS Template Settings > Apply DNS Template Changes.
And unless I manually click "Apply DNS Template Changes" for EVERY SITE it doesn't go away.
So, I was wondering, is there a way to scriptize that command? Or how to prevent it?
This is the command I use to generate DKIM and add it to Plesk database, also if any such record exists, delete and re-write it:
Code:
#!/bin/sh
ADMIN_PASS=`cat /etc/psa/.psa.shadow`
MYSQL_BIN_D=`grep MYSQL_BIN_D /etc/psa/psa.conf | awk '{print $2}'`
mysql="${MYSQL_BIN_D}/mysql -N -uadmin -p${ADMIN_PASS} psa"
query="select name from domains;"
domains=`echo $query | $mysql `
for i in ${domains}; do
echo "Processing $i"
[Code] ....
Is there any way for a client to use command line mail in a script. Client has ssh access via chromed environment.
Wants their script to be able to send an email on completion.
plesk 12 on Centos6.6
I have a subscription and I'm able to switch the subscription to a given service-plan, but I'm unable to sync the subscription with the plan, using the CLI. These are my commands:
/usr/local/psa/bin/subscription --switch-subscription <domain.com> -service-plan Hosting
result => The subscription <domain.com> is now associated with the service plan Hosting..When watching the webinterface in Plesk, I see the domain is "connected" to the service plan, so far so good. However, I notice that the subscription is not synced.
Back to the CLI, I've tried some combinations, but I can't get the subscription synced. The webbased interface keeps telling me that the subscription is unsynced. Here is what I've tried:
/usr/local/psa/bin/subscription --sync-subscription <domain.com>
And I also unlocked it (just in case it was locked) like so: /usr/local/psa/bin/subscription --unlock-subscription <domain.com>
The funny part is that the CLI tells me: "Subscription <domain.com> was synchronized with the associated service plans." but unfortenately it's not if I look through the GUI. Only if I click the sync-option webbased the subscription get synced. we have to transfer over 1000 domains and need to sync these domains with a given service-plan.
Although running
>>
/usr/local/psa/bin/dns --info dev.sname.c002.aname.domain.com
>>
shows the 'ecz1' TXT record exists for "dev.sname.c002.aname.domain.com" :
...
ecz1.dev.sname.c002.aname.domain.com. TXT ecz1
[Code] ....
The issue can be caused by extra spaces in the TXT record:
Code:
[root@a10-52-143-28 ~]# /usr/local/psa/bin/dns --add dev.sname.c002.aname.domain.com -txt 'ecz1 ' -domain ecz1
SUCCESS: Creation of DNS record in Domain 'dev.sname.c002.aname.domain.com' complete.
[root@a10-52-143-28 ~]# /usr/local/psa/bin/dns --del dev.sname.c002.aname.domain.com -txt 'ecz1' -domain ecz1
Unable to del record in DNS server: unable to find DNS record
In the following example I have 3 such TXT records with 3, 5, 7 extra spaces and remove record with 3 spaces:
Code:
[root@a10-52-143-28 ~]# mysql -u admin -p`cat /etc/psa/.psa.shadow` psa -e "select host, val, (length(val) - length(rtrim(val))) as extraspaces from dns_recs where host='ecz1.dev.sname.c002.aname.domain.com.' and val like 'ecz1%'"
+---------------------------------------+-------------+-------------+
| host | val | extraspaces |
+---------------------------------------+-------------+-------------+
| ecz1.dev.sname.c002.aname.domain.com. | ecz1 | 5 |
| ecz1.dev.sname.c002.aname.domain.com. | ecz1 | 3 |
[Code] ...
Click to expand...
Click for original message
I am running Plesk 12 on CentOS 7 64bit. I have set up the server backup instead of the website backup as I assume that will backup all server settings, as well as the website? I am running Plesk Web Admin Edition.
I have the backup set to run and dump into Personal FTP, but it does not run. If I run the backup manually, everything works great, but the scheduled backups don't work.
Sometimes I get this error:
Domain: mydomain.com
Plesk entry point: https://myserver.local:8443/
Following error is occurred during scheduled backup process:
[Errno 17] File exists: '/usr/local/psa/PMM/logs/backup-2014-06-18-23-30-05-556'
but in the backup.log and in the pmmcli.log I don't found any error.
What does it mean?
I have the same problem again with plesk 11 [URL]
workaround
1) rpm -q --scripts psa-backup-manager | sed 1d | sh
2) /etc/init.d/crond restart
my error message with CentOS 5.8, Plesk 11.0.9 MU#4
sh: line 2315: syntax error near unexpected token `('
sh: line 2315: `preuninstall scriptlet (using /bin/sh):'
/etc/cron.d/plesk-backup-manager is created
why do not work after you upgrade the backup cronjobs?
Plesk Backup work when we make instant backup but not with scheduled backup.
There is nothing in the log files (pmmcli.log & migration.log).
How can we correct?
Plesk 12 / Centos
on one of our Plesk-Servers (Plesk 12.0.18 Update 34 on Debian 7.6) the scheduled backup stopped working. Scheduled Backup is active in Backup-Manager, but it's not executed.
View 2 Replies View RelatedI'm running: 12.0.18 Update #29
I'm having troubles getting the global scheduled backup task to work at Home > Tools & Settings >Backup Manager > Server Repository.
At a subscription the scheduled backup works how it should.
Home > Subscriptions > example.com > Websites & Domains > Server Repository
The problem is as follows: When I set the scheduled task at a specific time example 00:00 it does not run at all. But when I just do "Create a backup" it does work...
I've done [URL] .... and everything in here is configured how it should be.
This is my configuration for scheduled task:
Cronjob is placed in : /etc/cron.d/plesk-backup-manager-task
Also did run: rpm -q --scripts psa-backup-manager | sed 1d | sh
This problem is only in the global backup configuration.
I have an issue with my daily backup. I configured to have a daily backup on an external FTP server. Everything looks to work correctly but in the panel I see that the task is a 100% forever. Never finishing completely.. So the next day, the task is not starting.. I have to remove manually the task in backup manager screen.
here is the last lines of the log file (in /var/log/plesk/PMM/backup-2015-01-20-20-46-02-824)
[2015-01-20 21:33:53.108|18485] INFO: ENV[LANG]=en_US.UTF-8
[2015-01-20 21:33:53.108|18485] INFO: Executing utility: /bin/sh -c -e /bin/tar --create --file /usr/local/psa/PMM/tmp/backupznMOWN/daily__1501202046.tar --directory /tmp/repo_transport_tmp_Jalkfn/ --dereference --files-from /tmp/repo_transport_tmp_UTCFhv
[2015-01-20 21:41:12.528|18485] INFO: The utility executed with the return code 0
[Code] ....
We have a dedicated server, with chroot access enabled.
We are wanting to backup each site on this server, as well as all mysql dbs on a weekly basis.
Is it possible for us to run a cron job to do this? If so, what commands would we want to use?
We have a local FTP server here, so would like to ftp these backups this machine.
I know it all sounds a little vague, but I just need to know where to get started.
My Centos server running Plesk12 is running scheduled backups every sunday 3AM.
The backup is configured so that it's created as a multivolume backup with a volume size of 2047MB.
The backup is placed on my Personal FTP repo (another plesk12 server mounted with big storage).
The backup content is configured to backup server config and content (all).
The problem I have is when the backup is running I can see that it creates the volumes and stores it locally. After it send all the volumes the the external FTP repo it will delete the local (tmp) data. See my attached screenshot for storage health during backup.
Is this behaviour normal? This way we can never run a backup to an external FTP repo when our server passed 50% storage. Is it not normal to:
- Create a volume
- Send it to FTP repo
- Delete volume locally
- Repeat until done
I'm new to Parallels Panel. I use version 11.0.9. I want to backup mysql database daily. First of all, what is the best way for daily database backup in plesk. I'm trying to do this in Scheduled Tasks and I use mysqldump command although I'm not sure.
I chose the time and day first and then I switched on the task. I typed the following command to Command line.
mysqldump -uusername -ppassword databasename | gzip -9 >/httpdocs/mydatabase.sql.gz
or
mysqldump -uusername -ppassword databasename >/httpdocs/mydatabase.sql.gz
This created only a blank file. When I use this without gzip, nothing changes.
1- Is mysqldump right command for database backup?
2- Should I define full path for mysqldump, gzip and database? If so, how can I find out the full path of mysqldump, gzip and my database? Because I can't see their locations in panel.
3- I can't see any error message. There is not any log file in httpdocs folder. Where does the log file exist?
4- It is weird but should the username be "database user" or should I write "root" ?
Is there a way to run an additional task after the scheduled backup has completed?I want to copy the backup off-site. I can't use FTP, otherwise I'd use the built-in 'Personal FTP Repository' feature. (See Rsync, Amazon S3, Rackspace cloud files, etc)
I could just create a new scheduled task, but the backup takes an unpredictable amount of time, and the tasks need to run sequentially. Also, the 2nd task should only run if the backup succeeds.
I could disable the backups in the control panel and create a new scheduled task that does both the backup and the additional task. But then the functionality of the control panel's Backup Manager page is lost.
I have recently upgraded Plesk Panel from 11.0.5 windows to 12.0.18 MU#46 windows.
As checked that local plesk backup via GUI throwing error and is not working. On checking the event viewer I found below event during backup run.
Log Name: Plesk
Source: scheduled_backup.exe
Date: 5/23/2015 6:52:06 AM
Event ID: 1
Task Category: None
Level: Warning
[Code] ....
I have created a scheduled task as per my application guide as in the attached file. When i recieve the email it says the following:
: /var/www/vhosts/testdomain.co.uk/clients/admin/cron.php: No such file or directory
i have also tried appending:
php -q
wget -O (using the http to the location to cron.php)
Nothing seems to work.
I have transfered from cpanel and this is my first scheduled task and not getting anywhere with very little information on how to do this.
Im running:
Plesk 12
Linux
after creating a new cron job, when i try to open Scheduled Tasks i get this error :Internal error: DateTime::__construct(): Failed to parse time string (2015-05-28 0.12:0:37) at position 17 ): Unexpected character
Message DateTime::__construct(): Failed to parse time string (2015-05-28 0.12:0:37) at position 17 ): Unexpected character
File Helper.php
Line 143
Type Exception
Whenever a user in my Cloudlinux powered Plesk Panel 12 tries to use the scheduled tasks, this is the error that is emailed to them...
excel: couldn't exec `/usr/local/psa/bin/chrootsh'
excel: No such file or directory
Does not matter what the task is, same error every time.
Even something as simple as wget [URL] .....
Looking at a server with Plesk 12 installed there is no root user shown in tools & settings / scheduled tasks.
Login as root and crontab -l shows root has some tasks.
On a similar server with Plesk 11.5 installed can see root user shown in tools & settings / scheduled tasks.
plesk 11.5
centos 6
We set a couple of task with domain user but the cron is not executed.
We tried 2 kind of task to run every minute
/usr/bin/php /var/www/vhosts/domain.com/httpdocs/cron.php
and
mail -s "Hello" admin@domain.com
1) Why scheduled task are not executed?
2) Where the tasks are written in filesystem from plesk ? /etc/....?
I recently updated from 11.5 to 12. After the update, my simple "Websites & Domains->Scheduled Tasks" is no longer executing.
I verified that the task is still present in Plesk and enabled.
The simple task launches one of my PHP web pages to perform an action:
(set to once a day: Min=30, H=7, DM=*, M=*, DW=*)
wget --quiet --delete-after "http://www.MYWEBSITE.com/some/path/index.php?someparam=somevalue"
This cron job had been working fine for more than a year but stopped working after the update.
I have the following command to execute every 5 minutes this PHP file which is working correctly:
/usr/bin/wget -qq -O /dev/null http://url.com/script.php
How can I make it so only the system (scheduled task or the server itself) can access to the file to execute it. I tried placing CHMOD 700 to the PHP file but everyone is still able to access to it and executive it.
Plesk version: 11.5.30 Update #44
We have a VPS with Centos 5 and Plesk 12.0.18.
The CRON service is running and works correctly, the Scheduled tasks on the administration Plesk Panel (tools & Setting / Scheduled tasks / root user) works ok, but when a customer sets a new scheduled task from her domain panel, don't works.
The task is a simple WGET command.
Im trying to zip up an sqlite database backup.
I execute this command
Code:
zip ../d/db/backup.zip ../d/db/09-02-15.sqlite
backup.zip never appears. Instead, I get some random filename in the directory. Like ziOHokOw
If I try to zip a smaller file(last weeks backup) everything runs fine?
Code:
zip ../d/db/backup.zip ../d/db/09-02-08.sqlite
So the 134mb file zips fine, but the 200mb one seems like its failing and im left with some type of temporary file. I tried downloading the weird filename and unzipping it. It has partial info like directory structure, the filename but the actual file inside is corrupt.
What is the Command Line, hoe do I get to it, when I do how do I update Perl to 5.88
View 5 Replies View RelatedI'm problem with crone sender email. I changed the version of the panel from 11.5 to 12. After start the corne and sehedule task in one of users I recived email. The crone task starts from email.
For this example, I replaced the user name in Plesk Panel, which is assigned cron jobs on “support”.
This is the mail system at host servername.domainname.
I'm sorry to have to inform you that message could not be delivered to one or more recipients. It's attached below.
For further assistance, please send mail to postmaster.
If you do so, please include this problem report. You can delete your own text from the attached returned message.
The mail system
<-f@domainname> (expanded from <-f>): bad address syntax
<support@domainname>: Host or domain name not found. Name service
error for name=support.domainnametype=AAAA: Host not found
[Code] ....
