Mickael Salfati
Regular Pleskian
Hi IgorG and other,
Always about IP failover, I noticed a very annoying issue with Plesk 10.x (10.1 or 10.2) wich didn't exist in 9.x branch.
Let's take an example.
Server 1 - Debian 5 lenny (1 Ip address and 2 IP failover) 3 adresses configured in plesk (Server >> Adress IP) and my /etc/network/interfaces
Server 2 - Debian 5 lenny (1 Ip address and 2 IP failover) 3 adresses configured in plesk (Server >> Adress IP) and my /etc/network/interfaces
As you can see failover IP on the 2 servers are identical, normal way to proceed and keep same Ip for end-customer.
Now, time to launch a migration on server 2 to migrate domains on server 1:
- When IP mapping is requested I have:
88.13.14.15 >> 88.13.14.15
88.16.17.18 >> 88.16.17.18
until here all sounds Good.
- I click to launch the upgrade.
Result: Migration doesn't finish and is blocked to 100%. some hours after the normal migration delay, I check process on server and nothing is processing (migration, perl, psa process, rsync, etc.) all my data in /var/www/vhosts, all my databases, all my maiboxes (/var/qmail/mailnames) are there on the server 2 but migration is always blocked to 100% (and processing)
Now if I do exactly the same process but about Mapping Ip I do:
88.13.14.15 >> 89.10.11.12
88.16.17.18 >> 89.10.11.12
(You'll notice that 89.10.11.12 is the real IP of the new server)
All sound good and migration is done.
Finally, plesk 10.x lose his way with failover IPs. I never encountered this issue with plesk 9.x and I use since many years the same way to migrate my domains.
I check all possible logs (in /opt/psa/PMM, /var/log, etc.) and I did'nt find any important informations.
Thanks per advance.
Always about IP failover, I noticed a very annoying issue with Plesk 10.x (10.1 or 10.2) wich didn't exist in 9.x branch.
Let's take an example.
Server 1 - Debian 5 lenny (1 Ip address and 2 IP failover) 3 adresses configured in plesk (Server >> Adress IP) and my /etc/network/interfaces
Code:
# The primary network interface
auto eth0
iface eth0 inet static
address 88.10.11.12
netmask 255.255.255.0
network 88.10.11.0
broadcast 88.10.11.255
gateway 88.10.11.255
auto eth0:0
iface eth0:0 inet static
address 88.13.14.15
netmask 255.255.255.255
auto eth0:1
iface eth0: inet static
address 88.16.17.18
netmask 255.255.255.255
Server 2 - Debian 5 lenny (1 Ip address and 2 IP failover) 3 adresses configured in plesk (Server >> Adress IP) and my /etc/network/interfaces
Code:
# The primary network interface
auto eth0
iface eth0 inet static
address 89.10.11.12
netmask 255.255.255.0
network 89.10.11.0
broadcast 89.10.11.255
gateway 89.10.11.1
auto eth0:0
iface eth0:0 inet static
address 88.13.14.15
netmask 255.255.255.255
auto eth0:1
iface eth0: inet static
address 88.16.17.18
netmask 255.255.255.255
Now, time to launch a migration on server 2 to migrate domains on server 1:
- When IP mapping is requested I have:
88.13.14.15 >> 88.13.14.15
88.16.17.18 >> 88.16.17.18
until here all sounds Good.
- I click to launch the upgrade.
Result: Migration doesn't finish and is blocked to 100%. some hours after the normal migration delay, I check process on server and nothing is processing (migration, perl, psa process, rsync, etc.) all my data in /var/www/vhosts, all my databases, all my maiboxes (/var/qmail/mailnames) are there on the server 2 but migration is always blocked to 100% (and processing)
Now if I do exactly the same process but about Mapping Ip I do:
88.13.14.15 >> 89.10.11.12
88.16.17.18 >> 89.10.11.12
(You'll notice that 89.10.11.12 is the real IP of the new server)
All sound good and migration is done.
Finally, plesk 10.x lose his way with failover IPs. I never encountered this issue with plesk 9.x and I use since many years the same way to migrate my domains.
I check all possible logs (in /opt/psa/PMM, /var/log, etc.) and I did'nt find any important informations.
Thanks per advance.
Last edited: