A sad tale of ProSupport woe with ReadyNAS Pro Pioneer!

mdgm wrote:
Would still suggest keeping the Pro Pioneer for backups. Data stored on a single device is not backed up.

True but data you cannot get to after it is backed up is worse then having no backup at all!

mdgm wrote:
If you don't want to keep the Pro Pioneer you should be able to sell it either via the ReadyNAS Marketplace (http://www.readynas.com/forum/viewforum.php?f=33) or on a site such as eBay.

I fully expect I could pass this turd on to another for a decent price but would that really be fair to someone else?

If anything the possibility of using an ITX MB and the existing back-plane for a custom build is what I would want and if it dies in the process then I guess it was "the plan" from the storage gods!

Thanks!

So here is the update:

After dragging on and getting frustrated with a lack of progress and the latest suggestion being to try using it via iSCSI, I called a halt.

I have spent too much time and trouble to try and identify and resolve an issue that I admit is a fringe case. To that end I did not want to continue the root cause research at my site and requested a "come to Jesus meeting" on how this problem was going to be resolved. Important people on the call and actions planned.

0. Have Dev team review logs for anything missed - not expected but additional eyes don't hurt.
1. Ship new shiny unit in case issue is related to unfixed/fixable OS image. Manually "copy" device/share/user configurations and then Rsync data from unit A to Unit B.
2. Once comfortable that the data has been migrated, ship the bad unit off to be tortured in the lab until it is made to talk/reveal the issue.

ensign.fodder wrote:
I have spent too much time and trouble to try and identify and resolve an issue that I admit is a fringe case. To that end I did not want to continue the root cause research at my site and requested a "come to Jesus meeting" on how this problem was going to be resolved. Important people on the call and actions planned.

The major takeaway was that NetGear wanted the next "incident" to take place in their lab. Well, as of 20 minutes ago that goal has failed.

ensign.fodder wrote:
0. Have Dev team review logs for anything missed - not expected but additional eyes don't hurt.

No idea if this happened or if anything was discovered.

ensign.fodder wrote:
1. Ship new shiny unit in case issue is related to unfixed/fixable OS image. Manually "copy" device/share/user configurations and then Rsync data from unit A to Unit B.

So after six days of nothing I emailed to check and they had the chassis but were waiting on drives to fill it with. Sounds like my upgrade from Pro Pioneer to Pro Business might not be present on new chassis.

ensign.fodder wrote:
2. Once comfortable that the data has been migrated, ship the bad unit off to be tortured in the lab until it is made to talk/reveal the issue.

Don't know if the unit will ever give up its secrets but at this stage of the game I just want to be f*cking done with the ReadyNAS and NetGear.

We are sorry the problem happen before we get everything in place. We knew it is a possibility but unfortunately couldn’t beat the clock.
We will get the new unit to your as soon as possible and thank you for sharing the log with us from this instance

So to update and possible close this thread.

Finally received new Pro 6 unit and drives, backup jobs to move shares and contents, confirmed file counts and sizes, shipped old unit off, and registered new unit 6/6/2012.

Results of torture testing in lab:

The out of memory condition has been reproduced on the unit, although the unit itself has not yet completely locked up. Regardless, a developer is looking into the out of memory condition to see where it is coming from.

It appears the method the software uses to read and write data via CIFS was causing a large amount of files’ extended attributes to be read and re-read very quickly, each attempt of which in turn attempts to allocate large amounts of physically contiguous memory for the operation. We’ve implemented a change when this type of behavior occurs that will fall back to a virtual memory allocation system that should be better for this type of operation. It should be noted that the Pro system I have in the lab has been throwing these allocation errors without hanging during my tests. While I do think that this is the source of the original issue, it is possible that the hangs are due to something else. I will be testing the code that has this change in it to verify that the allocation errors stop and that the unit remains stable.

It looks like everything was OK after a little more than a week of heavy CIFS load, part of which came from backup software. I believe that the change made in the firmware to allow for backup software's specific type of access pattern will be implemented in our next firmware release.

So the new unit got the mentioned firmware a while back and I re-pointed the backup software at it and no issues on the ReadyNAS side since. For my existing advanced support contract they issued a new one, but that is another fight since:
1. They originally said they could not transfer
2. So I asked for new
3. Then they could transfer but not extend the term
4. So finally I made a big enough stink that they decided to issue new contact

But in the f*cking around in trying to transfer the old my warranty has components expired in the future.

Standard Product Warranty:
 - 	Hardware Warranty:                       Available till 29/May/2017
 - 	Free Installation Support Warranty:      Expired on 28/Aug/2012
 - 	Power Supply Warranty:                   Expired on 29/May/2017
 - 	Accessories Warranty:                    Expired on 29/May/2017
 
Advanced Support / Hardware Contracts:
   	OnCall 24x7 3 Yrs, Phone+NBD Replacement, Cat. 2 - Available till 30/May/2015

So in summation, if you can reach the right people in the right places and you are not willing to let yourself be mistreated, mountains can be moved.

Just ask yourself before you buy, should that be the way it is or should you be treated better?