Notifications
Clear all
Functionality Issues
105
Posts
7
Users
2
Reactions
14.2 K
Views
@softraid-support I was running a Metashape 3D reconstruction project yesterday. It ran for 26+ hours until it eventually crashed the system. It was being run with files from my desktop, not from either Softraid internal drive. The MacPro fans were full blast on and off all day. Do you think this is a heat issue? (I received no warnings from the graphic cards)
Posted : 07/12/2023 11:51 am
Is this normal for Metashape to overload the CPU's? (the fans on the Mac Pro rarely go full blast) I looked in their forums but did not see anything. Maybe post there?
https://agisoft.freshdesk.com/support/solutions/
It is possible that overloading the graphics card is causing it to crash. But the fans on the mac Pro are very powerful at moving air. So while it could be the cards crashing, but probably not from heat.
https://appleinsider.com/articles/19/12/11/apple-engineers-reveal-how-they-prevent-mac-pro-overheat
I was looking around and saw a comment about connecting the XDR to the other card and see if it reduced crashes. (this was for Adobe products) and it often works. Might be worth a try.
Posted : 07/12/2023 2:21 pm
Rejoice! I just updated to Sonoma 14.2 and SoftRaid 7.6.1 and SoftRAID now runs without error/shutting down and recognizes my SoftRAID RAID named "Media", a four-disk internal RAID 5.
Thank You!
Screenshot and report are attached.
Posted : 12/12/2023 12:33 am
There were two separate issues, which is why it took so long. One was a configuration issue in the mac Pro. the other is third party utilities causing damage to MacOS, blocking SoftRAID components from loading. We are going to be dealing with the latter for a while, i think.
7.6.1 is officially released tomorrow, even though we posted it today.
Posted : 12/12/2023 12:51 am
If the disk set is raid 5 and only one disk shows I/O errors can that disk be removed from the raid and the parity be used to rebuild the data?
Posted : 12/12/2023 9:56 pm
Does either volume mount?
Save a SoftRAID tech support file and attach it. Do not do anything in the meantime.
Posted : 13/12/2023 11:21 pm
Any update? Any hope of mounting these drives?
Disk Drill recovered many of the files but many are corrupt. Archives that don't open, many images that are half missing, and video clips that are now just single frames.
Posted : 19/12/2023 12:44 pm
that is tough news. Can you post a small jpg that is damaged (but opens)?
Posted : 19/12/2023 3:00 pm
Movie files look the same but are only one frame.
Posted : 22/12/2023 2:17 pm
Dec 27 08:43:23 - SoftRAID Tool: I ran verification on the drive showing the error in this raid: "The command 94 encountered the error -54." Is there any hope in getting this drive mounted?
Posted : 27/12/2023 9:20 am
It appears the forum was damaged for a few days, I did not get any posts for moderation, sorry.
These screen shots are awful. It appears that parts of the volume are seriously damaged. What is strange, is some of the pics, like the #2 above (upper right) is OK. What size is that jpg?
Send a new support file and I can look at that error. that was when you went to validate the volume?
Posted : 27/12/2023 8:57 pm
The image with the green strip at the top is a Blender render and 377k. (path screenshot below)
When I search for .jpg files and preview through the files many don't have icons.
Many are only partial frames, and many are OK but recovered out of the folder structure.
This is brutal as I do research and development and thousands of files look nearly the same with some camera adjustment applied to one set VS another, or A/B render differences. Time stamps help but on any given day I may render dozens of frames titled 001.jpg.
I also run a 60-camera array system that names the 60 cameras 001-060. The original folder names are unique, but there are a thousand files called 023.jpg. I know that sounds like a dumb plan for file naming but there is post-processing stabilization specific to every "camera 23" image that needs to be batched applied. Anyway, this has been VERY stressful.
I thought Level5 Raid was safe?
I need a plan.
If the data is toast, I need to reformat and organize the scavenged files back to the RAID.
If it is savable I need to get on with saving it.
Please advise. Do these drives need to be replaced? (Two 8T Accelsior 4m2)
Posted : 28/12/2023 2:19 pm
@davo I can't say I have ever seen anything like this, especially from a simple crash. the APFS directory structure must have been badly damaged.
i don't see any evidence this is a hardware issue. What a disaster.
You can try R-Studio, but I would guess will have the same results.
I have only slightly better tools than you, so it would do no good to send it to me. Drive Savers can perhaps recover better, but I am afraid with this scenario, even they won't be able to, as APFS is much harder to recover from, as all files are deeply fragmented by design.
So this appears to be a start over scenario. I would be happy to take a look myself, but it may just delay your getting back on your feet.
Posted : 28/12/2023 6:42 pm
