LSFS Updated

[glecit]

1) Looks like snapshot based sync (well, VERY close...) so good news we've re-worked this code path in an upcoming post-fix.

2) We'd be happy to take a look over remote session as I'm still not 100% sure about 1) so please drop a message to support with a link to this thread and credentials for you RDP or whatever (TeamViewer?) account.

Thank you for cooperation!

It's stuck during a fast sync again after a clean reboot. I'm going to leave it like this for now in case you want to do a remote session while it's stuck.

It would be nice to hear if this is a known issue or not.

Sorry, after a few days I needed to get the lab cluster running again to test something else. I can provide logs if you're interested, but it's probably not worth the time to debug old code.

Looking forward to testing the new sync code. By "VERY close" did you mean to a new beta build or to a production release?

Thanks, Anton.

[anton (staff)]

We're getting closer so probably you're right and it worth moving to new build eventually

Beta but it would be released in a shorter cycle as we did not change core (with the current update, tons of changes in the trunk for another release) dramatically.

[wgeddes]

I never did give an update, but here is one now...

I created 3 separate lsfs disks. 2 of them 50gb each (for hyper-V OS), and 1 x 300GB (Shared SQL Data).

Actual space being used in the lsfs disks:
1st 50GB - 33GB Used Space
2nd 50GB - 33GB Used Space
300GB - 5GB Used

Actual space the Starwind lsfs disks are using:
1st 50GB - 426.5GB
2nd 50GB - 426.38GB
300GB - 1.42TB

Starwind is using 2.28TB of physical space with 400GB lsfs drives in total.

After creating the disks on the 1st node, I let Starwind create the replica on the 2nd without altering any options. Everything showed fine on setup. Now when logging in, the 2 x 50GB disks arent even showing in the console for the 2nd node, so yellow triangles for both of the HA disks on the 1st node, and the SQL shared 300GB disks shows partner not synchronized.

I did also create a 1GB fixed disks for a cluster witness, and this disk is performing as it should.

[Anatoly (staff)]

Starwind is using 2.28TB of physical space with 400GB lsfs drives in total.

So 5 times more. Still not good, but already better. We are on the right way!

After creating the disks on the 1st node, I let Starwind create the replica on the 2nd without altering any options. Everything showed fine on setup. Now when logging in, the 2 x 50GB disks arent even showing in the console for the 2nd node, so yellow triangles for both of the HA disks on the 1st node, and the SQL shared 300GB disks shows partner not synchronized.

Any chance you could upload the logs and share the link to download them with us?

[fbifido]

Hi Anatoly (staff),

can you post a link to the latest version 6 ?

Thanks.

[Anatoly (staff)]

I`m pretty sure that you are referring to version 8 (correct me if I`m wrong). It can be downloaded from the customers page:
www.starwindsoftware.com/customer-page

[fbifido]

I`m pretty sure that you are referring to version 8 (correct me if I`m wrong). It can be downloaded from the customers page:
http://www.starwindsoftware.com/customer-page

No, not v8. What I am asking about is v6
as in the previous version.

Thanks.

[rrbnc]

Is this fixed in v8 build 7774?

https://www.starwindsoftware.com/release-notes-build

"LSFS

Performance optimizations.

Defragmentation issues fixed. In some cases, defragmentation process could stop, and device files take more disk space then required. Currently during the normal operation, device files can take disk space up to 3.5 times more compared to nominal device size. Snapshots increase disk space requirements by unique data size in snapshot.

Deduplication option temporarily disabled for new devices."

[anton (staff)]

Yes, that's the most recent build with an updates.

[jamesalan]

Raising the priority of the defrag process to manage disk space more efficiently also sounds like a critical enhancement. Keen to see how this update impacts performance and reliability in real-world setups!

[darklight]

Raising the priority of the defrag process to manage disk space more efficiently also sounds like a critical enhancement. Keen to see how this update impacts performance and reliability in real-world setups!

Are you still running LSFS in your environment? Could you be so kind as to share details about your workload and experience?

[anton (staff)]

LSFS shouldn't be used in production anymore. We pulled it out looooong time ago!