ADSM-L

[ADSM-L] SV: [ADSM-L] Thinking about replicate node command RFE, feedback welcome

2017-02-01 06:51:56
Subject: [ADSM-L] SV: [ADSM-L] Thinking about replicate node command RFE, feedback welcome
From: Conny Landstedt <conny.landstedt AT SLL DOT SE>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Wed, 1 Feb 2017 11:48:17 +0000
I would vote for that RFE, I see the same issue with our setup.

Adding an option like wherereplstate=SEND,SENDSYNC would solve this issue.

Best regards
Conny Landstedt

> Do you guys think that would be worth an RFE?

I think so, yes.
We're struggling with the same problem; for now we were just ignoring those 
ANR2728E messages, because avoiding them would make our daily maintenance 
script way to complicated.
We have an open PMR at the moment where IBM support is asking us the following:

---
I noticed in the ACT LOG, there are a lot of ANR2728E.

ANR2728E REPLICATE NODE: Node CLGHV001_DBGHV001PROD       
                          cannot be replicated because it is not in the SEND or 
   
                          SYNCSEND mode. (SESSION: 2937994, PROCESS: 5406)

Are these nodes supposed to be replicated? If so, please have the customer 
correct the issue.
Otherwise, I recommend the customer remove them from their replication. This 
will keep things clean and not add any extra noise as we analyze the problem.
---

Yes, well, if there were an elegant way of removing them from replication, we 
would.

Two servers doing cross replication seems like an obvious setup, where you will 
always bump into this problem.
A replicate node option like you describe would help us out and apparently 
answers to a need of IBM support as well.

Regards,
Frederik de Kezel

-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of 
Stefan Folkerts
Sent: maandag 30 januari 2017 09:50
To: ADSM-L AT VM.MARIST DOT EDU
Subject: [ADSM-L] Thinking about replicate node command RFE, feedback welcome

Hi all,

I'm thinking about creating an RFE for the replicate node command but I was 
wondering if I might be missing something so obvious that my RFE would be 
nonsense.

I try to create setups without and errors or warnings from housekeeping scripts.
I also try to work with opt-out mechanisms as much as possible for customers 
that agree to use that backup philosophy.

When you build a two-way replicating Spectrum Protect setup using node 
replication (with or without protect stgpool) you get a LOT of errors of this 
type.

ANR2728E REPLICATE NODE: Node <Nodename> cannot be replicated because it is not 
in the SEND or SENDSYNC mode.

You either have to use nodegroups, single nodes or accept these messages, this 
doesn't work with an opt-out policy.
Why isn't there a replicate node option like with move drmedia where you can 
say something like wherereplstate=SEND,SENDSYNC.

Do you guy's think that would be worth an RFE?

Regards,
   Stefan
<Prev in Thread] Current Thread [Next in Thread>