On 3/31/2021 at 5:23 PM, Joseph said:

 Unfortunately, that command doesn't seem to be working... I'll let you know after I get it up and running and try the command again.

 

So, the backup box is back up. The 10Gbe cards I just installed don't seem to be making much of a difference, but perhaps it's just my perception. fwiw, I'm not using the cache drive or SSD for TM it's the only thing using the backup unit for now. (Results from the command are attached)

 

8.21 MB/s is terrible!! I have no idea what's going on. I stand by my original observation: SMB transfers are much faster after unchecking "Put hard disks to sleep when possible" in the Energy Saver settings than it used to be; but it seems I should be getting faster R/W speeds with the new 10GBe NICs, so I suspect the new NICs aren't configured properly... idk. ...guess that's a topic for another thread. 

 

ps. not sure why the error at the end of the log: error: Disk object invalid or unable to serialize

 

Here's a screenshot of TM while running. It went from 1 hour to 2 hours remaining as I'm typing...wth 🙁

results.txt

Edited April 3, 20215 yr by Joseph
updated results as attachment

19 hours ago, Joseph said:

8.21 MB/s is terrible!! I have no idea what's going on.

 

That's what I'm getting at. We're blaming unRAID but I think Time Machine itself's the limiting factor.

 

You can disable Time Machine throttling by running this on your Mac which might speed it up (it gets reset on reboot)

sudo sysctl debug.lowpri_throttle_enabled=0

 

Still that log showing a 40GB backup in about an hour and a half isn't bad, better than what I get so it might be a combination of factors. Is there a good reason the backup was so large? Mine are usually around 1GB.

 

Not sure about your "Recovery" error.

Edited April 3, 20215 yr by CS01-HS

2 hours ago, CS01-HS said:

 

That's what I'm getting at. We're blaming unRAID but I think Time Machine itself's the limiting factor.

agreed.

 

Quote

You can disable Time Machine throttling by running this on your Mac which might speed it up (it gets reset on reboot)

I'll give that a shot next time I can and let you know how it goes

 

Quote

Still that log showing a 40GB backup in about an hour and a half isn't bad, better than what I get so it might be a combination of factors. Is there a good reason the backup was so large?

Agreed, but it wasn't like that until I prevented the disks from sleeping; but like you said, it could be a combo of things. TM backups are typically 7G+ and I don't have a clue why... wasn't like that on my old Mac when I had TM as a local drive. I haven't used TM in awhile stuff was moved around which could perhaps explain this 40G session, but the large backups in general are a mystery.

 

Quote

Not sure about your "Recovery" error.

I'll dig around and see what I can find. But it may, along with the other issues, require the use a few 'lifelines' to the unRAID community in other threads

Maybe it's backing up the full Recovery partition every time?

 

Stab in the dark but I believe spotlight indexing of the mac and the backup plays a role.

You can rebuild the mac index easily enough by turning it off:

sudo mdutil -a -i off

then on:

sudo mdutil -a -i on

but the backups are more complicated, maybe easier to start fresh.

You can set multiple time machine destinations so you wouldn't have to wipe the old one.
 

if you want the best TM backups you need to use a full SSD drive for the backup share and use AFP. Never use user share with cache enabled on the backups as it creates much more overhead. Just use prefer or no cache.

spotline indexing on the TM drive cannot be controlled much so if you happen to hit the time right when the indexing is running you are screwed. therefore the SSD is the only solution here. its because the sparse bundle is buil tof thousands of files which need to grow

Edited April 11, 20215 yr by theruck