1&1 Dedicated Server serial backup/restore

[Funboy-63]

Hi to all,

Well all I can say it's defiantly Friday the 13th for me!

In the early hours of this morning I had a server hardware fault which has resulted in my entire server being lost, not to worry I thought, I have my backups on the 1and1 backup server which is cron'd by Plesk, well that's all well and good until I tried to restore them.

First I have imaged my server to the same Centos 5 version which included Plesk 10.0.1, then via SSH I incrementally upgraded to each version to ensure every thing was reinstalled up to the latest version I had this being 10.2.0 as I have had issue before when trying to jump straight to the latest version.

Now the fun starts as if I try to do a restore in Plesk which requires me to copy the personal FTP (1and1 backup server) repository tar file to the server repository all hell breaks lose resulting in below error or even a server 500 error with no access to Plesk CP!

I had this once before many years ago and remember there was a way to restore the 1and1 backup server tar files via the serial connection you get with 1and1 accounts trouble is I cannot for the life of me remember the commands need to accomplish this so wondered in anyone here would know and care to share it please.

I am looking to restore a backup of 4.80GB which might be why Plesk is having a fit, alternatively anyone know the command to download it locally for archive as I cannot afford to lose it. Again if I try this in Plesk it results in errors.

Here is the error being received while in Plesk when trying to do a transfer or download from the personal FTP repository.

Error: pmm-ras error (Error code = -6):
== STDOUT ====================

==============================
== STDERR ====================
terminate called after throwing an instance of 'plesk::CurlException'

==============================

Here's hoping some bright spark can come to my rescue.

 

[IgorG]

These connections are issued by Plesk Backup Manger - whenever remote FTP repository is browsed through Control Panel interface, Backup Manager downloads all XML descriptions for each backup in the directory. It appears that each file is downloaded in its own connection instance which causes FTP server to promptly reject such connections.
The only solution I can suggest is to increase limits on number of connections on FTP server.

 

[Funboy-63]

Hi to all,

If anyone is getting the following error:

Error: pmm-ras error (Error code = -6):
== STDOUT ====================

==============================
== STDERR ====================
terminate called after throwing an instance of 'plesk::CurlException'

==============================

You might want to check your tmp size, when I had the above error my tmp partition was only 2GB in size while the backup file I was restoring was 4.80GB. So all I did was increase my tmp partition size and all went smoothly.

HOW TO INCREASE YOUR TMP PARTITON (NOTE: This following tut info is not mine and I cannot remember the original source as I have had it for a while now and shows you how to create a 2GB partition so increase to your desired size)

How to increase the size of /tmp partition in Linux


Go ahead with the given steps:-

1)Stop MySql service and process kill the tailwatchd process.

[root@localhost ~]# /etc/init.d/mysqld stop
Stopping MySQL: [ OK ]
[root@localhost ~]#pstree -p | grep tailwatchd
Find the tailwatchd process id and kill it
[root@localhost ~]# kill -9 2522

# Use the below command if pstree -p | grep tailwatchd fails to work on your system, this is what I had to do. (Funboy-63)
=======
lsof |grep /tmp
=======

You'll get the process ID and kill the processes like kill -9 PID. Here is an example(Second column is PID)

2)Take a backup of /tmp as /tmp.bak
[root@localhost ~]#cp -prf /tmp /tmp.bak

3)Create a 2GB file in the available freespace
[root@localhost dev]# dd if=/dev/zero of=/usr/tmpDSK bs=1024k count=2048
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB) copied, 73.6908 seconds, 29.1 MB/s

[root@localhost dev]# du -sch /usr/tmpDSK
2.1G /usr/tmpDSK
2.1G total

4)Assign ext3 filesystem to the file

[root@localhost dev]# mkfs -t ext3 /usr/tmpDSK
mke2fs 1.39 (29-May-2006)
/usr/tmpDSK is not a block special device.
Proceed anyway? (y,n) y
Filesystem label=
OS type: Linux
262144 inodes, 524288 blocks
26214 blocks (5.00%) reserved for the super user
First data block=0
Writing inode tables: done
Creating journal (16384 blocks): done
Writing superblocks and filesystem accounting information: done
This filesystem will be automatically checked every 25 mounts or
180 days, whichever comes first. Use tune2fs -c or -i to override.

5) Check the file system type:-

[root@localhost dev]# file /usr/tmpDSK

/usr/tmpDSK: Linux rev 1.0 ext3 filesystem data (large files)

Note:-

You may also use the following comands for making ext3 file system on a file:
[root@localhost dev]#mkfs.ext3 /usr/tmpDSK
[root@localhost dev]#mke2fs /usr/tmpDSK

6)Unmount /tmp partition

[root@localhost dev]#umount /tmp

7))Mount the new /tmp filesystem with noexec

[root@localhost dev]#mount -o loop,noexec,nosuid,rw /usr/tmpDSK /tmp

Set the correct permission for /tmp

[root@localhost dev]# install -d –mode=1777 /tmp
[root@localhost dev]# ls -ld /tmp
drwxrwxrwt 3 root root 4096 Feb 6 08:42 /tmp
( you may also use the command chmod 1777 /tmp for doing the same )

[root@localhost dev]# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda9 28G 6.4G 20G 25% /
/dev/sda8 99M 10M 84M 11% /boot
tmpfs 500M 0 500M 0% /dev/shm
/usr/tmpDSK 2.0G 68M 1.9G 4% /tmp

8)Restore the content of old /tmp.bkp directory

[root@localhost dev]# cp -rpf /tmp.bak/* /tmp

Restart the mysql and tailwathchd services.

[root@localhost dev]# /etc/init.d/mysql start
[root@localhost dev]# /scripts/restartsrv_tailwatchd

9)Edit the fstab and replace /tmp entry line with :-

/usr/tmpDSK /tmp ext3 loop,noexec,nosuid,rw 0 0

10) Mount all filesystems

[root@localhost dev]#mount -a

Check it now:-

[root@localhost ~]# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda9 28G 6.4G 20G 25% /
/dev/sda8 99M 10M 84M 11% /boot
tmpfs 500M 0 500M 0% /dev/shm
/usr/tmpDSK 2.0G 68M 1.9G 4% /tmp

Done.

 

[fnoob]

Didnt work for me - I cant access the plesk panel any more.

 

[BillSTW]

instead of making your /tmp partition larger - just tell plesk to use some other location

http://kb.parallels.com/en/6136