Help please with learning about Replication

Status
Not open for further replies.

Redcoat

MVP
Joined
Feb 18, 2014
Messages
2,925
My first experience with replication. I've manually set up replication between my FreeNASMini (PUSH) and my Dell C2100 (PULL), presently sitting physically side by side. On the PUSH side, GUI Replication Status indicates "Up to date". On the PULL side GUI under "Snapshots" I can find all the PUSH side's snapshots listed including that listed as the "last sent". What I do not have on the PUSH side GUI are any entries in the the "Snapshot" "Replication" column where I understood I would see some indication that Replication of the row's particular snapshot had taken place. Can someone explain what's going on here and, if anomalous, what I might check and how, please?
 

nojohnny101

Wizard
Joined
Dec 3, 2015
Messages
1,478
I have found the GUI to be a bit funky at times with what information it displays regarding replication.

Only way to test for sure if everything you want has transferred over is to clone a snapshot on PULL and then mount a share for it, then browse it. Go in and see if your files are there. I would open a couple of them for extra measure. I do this every so often even though replication has been running flawlessly for me for the past 6 months (knocks on wood).
 
Last edited:

Redcoat

MVP
Joined
Feb 18, 2014
Messages
2,925
I have found the GUI to be a bit funky at times with what information it displays regarding replication.

Only way to test for sure if everything you want has transferred over is to clone a snapshot on PULL and then mount a share for it, then browse it. Go in and see if your files are there. I would open a couple of extra measure. I do this every so often even though replication has been running flawlessly for me for the past 6 months (knocks on wood).
Thanks for the info - I will work through your suggestion and let you know what I find. I will be interested to see if anyone else confirms my experience specifically...
 

PhilipS

Contributor
Joined
May 10, 2016
Messages
179
I don't think I have ever seen anything in the replication column in Storage, Snapshots. I just look in Storage, Replication Tasks at the Last snapshot sent to the remote side.
 

Redcoat

MVP
Joined
Feb 18, 2014
Messages
2,925
I don't think I have ever seen anything in the replication column in Storage, Snapshots. I just look in Storage, Replication Tasks at the Last snapshot sent to the remote side.
Yes, thanks, I made the same x-reference. My expectation was created by the manual's: "Replication shows whether the snapshot has been replicated to a remote system."
 

wblock

Documentation Engineer
Joined
Nov 14, 2014
Messages
1,506
There is a known bug with the replication column status, it's being worked on now.

I rewrote the Replication section in the User Guide a little while back with an eye to simplifying it and yet showing good examples. It is intended to be clearer than the previous version: http://doc.freenas.org/9.10/storage.html#replication-tasks.
 

PhilipS

Contributor
Joined
May 10, 2016
Messages
179
There is a known bug with the replication column status, it's being worked on now.

I rewrote the Replication section in the User Guide a little while back with an eye to simplifying it and yet showing good examples. It is intended to be clearer than the previous version: http://doc.freenas.org/9.10/storage.html#replication-tasks.
I noticed the rewrite, thanks for that. Any reasoning for switching from the PUSH/PULL terminology to Alpha/Beta verses something like Source/Destination?
 

wblock

Documentation Engineer
Joined
Nov 14, 2014
Messages
1,506
"Push" and "pull" are confusing to read, because both are verbs. However, the "pull" system does not actually "pull" things, it just receives them. I considered other names also, but finally settled on names that would not be confused with verbs but also allow describing situations like replicating from one system to two others, or from one to another which then replicates to another. Using numbered names would be less clear, so went with the alpha/beta scheme.
 

Redcoat

MVP
Joined
Feb 18, 2014
Messages
2,925
There is a known bug with the replication column status, it's being worked on now.

I rewrote the Replication section in the User Guide a little while back with an eye to simplifying it and yet showing good examples. It is intended to be clearer than the previous version: http://doc.freenas.org/9.10/storage.html#replication-tasks.

Thanks for the acknowledgement of the bug. With respect to the rewrite, it did reflect the GUI changes but I haven't tried the automated task set up for replication so I cannot perhaps appreciate its elegance. I didn't find the old version tough to handle, actually.
 

Redcoat

MVP
Joined
Feb 18, 2014
Messages
2,925
Thanks for the acknowledgement of the bug. With respect to the rewrite, it did reflect the GUI changes but I haven't tried the automated task set up for replication so I cannot perhaps appreciate its elegance. I didn't find the old version tough to handle, actually.
Update. This morning the Replication column shows "OK" for all the snapshots. System changes yesterday were update to FreeNAS-9.10.2-U2 (e1497f2) and checking the box to save DDR logs in the system dataset (see thread 45998).
 

cbarber

Dabbler
Joined
Sep 23, 2017
Messages
17
"I have found the GUI to be a bit funky at times with what information it displays regarding replication."

That's the understatement of the century :)
 

cbarber

Dabbler
Joined
Sep 23, 2017
Messages
17
"Push" and "pull" are confusing to read, because both are verbs. However, the "pull" system does not actually "pull" things, it just receives them. I considered other names also, but finally settled on names that would not be confused with verbs but also allow describing situations like replicating from one system to two others, or from one to another which then replicates to another. Using numbered names would be less clear, so went with the alpha/beta scheme.

"Source" and "Destination" ???
 

wblock

Documentation Engineer
Joined
Nov 14, 2014
Messages
1,506
In the quoted section, I explained why that fails when you have more than one destination, either with one machine replicating to multiple others, or with one machine replicating to another, which then replicates to a third. It gets even worse when you consider mirror-like schemes, where things on machine A are replicated to machine B, and some things on machine B are replicated to machine A.
 
Status
Not open for further replies.
Top