SoftRAID (issues)

I have the same model disks in a test unit. I did have problems for a few days, but then it stabilized and works fine.

Your experience with DU is interesting. One idea to test is check the preference in the volumes tab to "submit IO's one at a time". restart.

See if that makes a difference. It should not, but that is the primary difference between Apple's RAID and SoftRAID, Apple RAID is single threaded. Both use the standard OS X tools to perform reads and writes.

Let us know.

Yes, DU seems to have issues initializing disks being "managed by" SR. We're going to stay with DU volume for now, and keep an eye on the errors through a utility. The sheer amount of error issues through SR is unbelievable. We have restarted our computer more times in the last 2 days than in the last 2 years.

We did just update to 5.7.3.
Will this help?

(we actually just updated SoftRAID, and did convert the volume from Disk Utility to SoftRAID. We'll see what happens... !)

Let us know. Also the "submit reads and writes one at a time" setting.

Yes, DU seems to have issues initializing disks being "managed by" SR. We're going to stay with DU volume for now, and keep an eye on the errors through a utility. The sheer amount of error issues through SR is unbelievable. We have restarted our computer more times in the last 2 days than in the last 2 years.

This DU initialize issues started around Mavericks, if I remember.

Sorry you are having problems with this enclosure and SoftRAID. Both SoftRAID and Disk Utility RAID use the "disktool" in OS X for reads and writes. DU RAID is single threaded, however, one of the main differences.

Just restarted computer, and ALL volumes and disks associated with SoftRAID have i/o Errors, which i don't believe is true. This software has been awful. I cannot even mount the 4M2E right now, and will be losing data. Seems like this software isn't really ready for primetime.

SoftRAID is just the messenger. It reports all disk errors (IO errors) regardless of cause. The pointer is your hardware is not stable at this time. OS X ignores all disk errors, unless they cause some form of immediate data loss. We believe it is better for users to be warned than be oblivious of problems.

The errors on this may have been caused by a shutdown panic. (something hung at shutdown)

Clear the IO errors in this case. Look to see if there is a crash log from your shutdown, and where the crash was located. It may give a clue.

SoftRAID has been stable for over 20 years. Our basic driver has passed Apple's stringent OS X tests for being bundled with OS X for over 15 years.

The OS X "disktool" tool actually performs all disk reads and writes, which prevents compatibility problems. It is the same method the Apple RAID driver works, by passing all read/write commands to the underlying OS X tool for completion.

We need to get to the bottom of your hardware issue. Thunderbolt is very fast, so even something minor can cause weird problems.

SoftRAID is just the messenger. It reports all disk errors (IO errors) regardless of cause. The pointer is your hardware is not stable at this time. OS X ignores all disk errors, unless they cause some form of immediate data loss. We believe it is better for users to be warned than be oblivious of problems.

The errors on this may have been caused by a shutdown panic. (something hung at shutdown)

Clear the IO errors in this case. Look to see if there is a crash log from your shutdown, and where the crash was located. It may give a clue.

SoftRAID has been stable for over 20 years. Our basic driver has passed Apple's stringent OS X tests for being bundled with OS X for over 15 years.

The OS X "disktool" tool actually performs all disk reads and writes, which prevents compatibility problems. It is the same method the Apple RAID driver works, by passing all read/write commands to the underlying OS X tool for completion.

We need to get to the bottom of your hardware issue. Thunderbolt is very fast, so even something minor can cause weird problems.

Thank you for the thoughtful reply.

The issue is the inconsistency of error reporting from SoftRAID. It's all over the place. When we use Drive Scope + Disk Utility, and have no reported errors/issues. On every restart of our Mac Pro, the program reports different disks as reporting errors. Today, SoftRAID reported every disk being managed by SoftRAID as reporting errors, and saying to replace disks as soon as possible. The created RAID 0 was unmounted by SoftRAID, and would not mount.

*Yet the disks are all fine*

I have once again re-initialized all disks in Disk Utility, and am reporting zero issues with Drive Scope / Tech Tools / Disk Warrior.

All the tools you are using can do is A) ask for a SMART test, and B) check SMART parameters (which are useless on SSD's at this point).

SoftRAID is giving you a real time report that the disks had an error.

There are many things that can cause an error, without the disk being faulty.
There can be hardware issues where cables, disks, or timing causes errors.
There can be directory damage where it asks the disk to read to a non existent location, which would cause an error.

A "disk error" in SoftRAID is telling you something is wrong. OS X does not give enough information to do more than that.

You can imagine driving a car normally with the check engine light on. The tires have air in them, and the radiator is full, but neither tells you why the light is on. And a check engine light can be triggered by something as banal as leaving the gas cap open when driving.

Same with your system. You know there is an issue. Next step is eliminating the cause of the problem.

I mentioned the cause of the current errors was probably a hang at shutdown. Clear the errors. Again, SoftRAID told you there is an issue, but not the why. The reason SoftRAID can tell you, is the disk is in SoftRAID format, and the SoftRAID driver tracks and reports ALL disk error events.

Disk Utility ignores them. So you can be getting the same kinds of errors and never notice, until you discover file corruption down the road.

So what is a person to do?
This is a brand new setup.

Physically reset the blades in the enclosure, make sure they are tight.
Certify the blades again.
Make sure your cables are tightly connected
(if you have a spare TB cable, try replacing it)

I do not have a sure fire way to fix this for you. I have seen this issue in house, but now that my Samsing blades "stabilized", there are no longer any issues with the M2/4 enclosure with these blades in it.

Perhaps it was a second certify that did it, it may have been re-installing the blades, I just don't know. We would be happy to give you more information if we could.

The enclosure is just a pass-through, pretty basic. I don't think that is triggering these issues. Especially when the main reports I hear are only on Samsung blades, rarely on any other. (Although I confess this could easily be either sample size, or the fact that far more users purchase Samsung blades than any other brands.)

One problem enclosure I was supporting turned out to be a defective blade. It took a second or third certify to figure that out.

Okay, thanks, will report back.

Reseated the drives.
Switched TB cable
Converted volume to SoftRAID
Certifying all 4 disks again (previous Certify reported no issues).

*Also, the 4M2E is daisy chained through a Pegasus3. They are the only 2 units on this TB bus.