Making a first build

Status
Not open for further replies.

Matdif

Explorer
Joined
Oct 10, 2014
Messages
59
Redid my cables used different sata ports. While looking around the forum I am noticing lots of people having trouble with this Marvel controller for half the sata ports. Im going to run badblocks again and pay real close attention to which sata ports are throwing the errors. If there is an actual solution to this marvel problem let me know itll be a couple days or more till this badblocks test is done.
 

Matdif

Explorer
Joined
Oct 10, 2014
Messages
59
5 1/4 hours in no errors all nine drives 0/0/0. It looks like this thing fixed itself somehow. Has this happened for other people?

Edit: Continuing to badblocks but I am noticing these things are moving at different speeds. I started them all within 5 mins of each other. Now they are all within 1-2% of each other but the order is out of whack.

How do you tell if the speed difference is too much like jgreco mentions here. It might be elsewhere but he mentions too much difference in read write speeds being a sign of a bad disk and should be stress tested.

https://forums.freenas.org/index.php?threads/building-burn-in-and-testing-your-freenas-system.17750/
 
Last edited:

cyberjock

Inactive Account
Joined
Mar 25, 2012
Messages
19,526
When jgreco says the speed difference is too much he means you are expecting 100MB/sec and instead getting 5MB/sec. You'll know if the speed difference is too much because there will be red flags going off in your head saying something is wrong.
 

Matdif

Explorer
Joined
Oct 10, 2014
Messages
59
Ah ok thanks. 13 hours and everything seems fine, has not thrown an error yet.

Edit: Whelp I was wrong. I made a folder in my email for freenas stuff and that somehow made my phone not pick them up or alert me. Found errors when I logged into my email this morning on a computer.

Well I am still getting the errors. Thought I was free but I was wrong.

These were from a week and a half ago or so on my first badblocks test.

Device: /dev/ada4, Read SMART Self-Test Log Failed
Device: /dev/ada1, Read SMART Error Log Failed
Device: /dev/ada2, Read SMART Self-Test Log Failed
Device: /dev/ada4, Read SMART Error Log Failed
Device: /dev/ada3, Read SMART Self-Test Log Failed
Device: /dev/ada3, Read SMART Error Log Failed
Device: /dev/ada4, unable to open device
Device: /dev/ada3, failed to read SMART Attribute Data
Device: /dev/ada1, Read SMART Self-Test Log Failed
Device: /dev/ada3, Read SMART Self-Test Log Failed
Device: /dev/ada1, failed to read SMART Attribute Data
Device: /dev/ada1, failed to read SMART Attribute Data
Device: /dev/ada1, failed to read SMART Attribute Data
Device: /dev/ada1, unable to open device
Device: /dev/ada3, ATA error count increased from 0 to 1

I started again last night and this is what I have so far. In between these two I redid cable management and switched cables and sata slots.

Device: /dev/ada3, Read SMART Error Log Failed
Device: /dev/ada2, failed to read SMART Attribute Data
Device: /dev/ada2, Read SMART Self-Test Log Failed
Device: /dev/ada3, Read SMART Self-Test Log Failed
Device: /dev/ada3, failed to read SMART Attribute Data
Device: /dev/ada3, failed to read SMART Attribute Data
Device: /dev/ada2, Read SMART Self-Test Log Failed
Device: /dev/ada3, Read SMART Self-Test Log Failed
Device: /dev/ada2, Read SMART Error Log Failed
Device: /dev/ada2, failed to read SMART Attribute Data
Device: /dev/ada3, failed to read SMART Attribute Data
Device: /dev/ada2, failed to read SMART Attribute Data
Device: /dev/ada3, Read SMART Self-Test Log Failed
Device: /dev/ada3, Read SMART Self-Test Log Failed
Device: /dev/ada2, failed to read SMART Attribute Data
Device: /dev/ada2, Read SMART Self-Test Log Failed

Looks like its going to be the maxwell card but I havent traced the sata slots yet.

edit2: putty reporting no errors.

putty.jpg
 
Last edited:

Matdif

Explorer
Joined
Oct 10, 2014
Messages
59
Been doing badblocks and re-seating sata cables from drives that throw errors. Switched to thinner cables that dont push against each other when plugged into the motherboard. I am noticing that when some drives error its always the Marvell ones but not all of them, the other drives on the Marvell dont error. I had 2 drives do it last night at midnight then nothing for seven hours. I reseated those sata cables in other spots and restarted the badblocks test. Nothing so far for 2 hours when in the beginning the errors came every 15 mins.

I am going to let this test run its full course to see if it produces any erors should take 3 or more days as these drive are 4tb. Anyone know a better test for pushing these Marvell sata ports? I dont want this to be a problem that could creep up anytime after I make a pool and start holding data on this.
 

Fraoch

Patron
Joined
Aug 14, 2014
Messages
395

Matdif

Explorer
Joined
Oct 10, 2014
Messages
59
iozone has been kicking the **** out of my pool for more than a week now. It's that last punishing test located here:

https://forums.freenas.org/index.php?threads/checking-new-hdds-in-raid.12082/#post-98457

Drives are still holding up fine, touch wood. But that test takes a very long time and stresses the drives continuously.

Thanks Ill look into this after. I got time till my satadom shows up so I can take it slow. Might move all the drives onto the Marvell as well for a few days to push it.
 

Fraoch

Patron
Joined
Aug 14, 2014
Messages
395
The first few iozone tests don't take that long. It's this last one that's been going on and on and on. The reads/writes seem minor, 100s of kB/s or a few MB/s, but they are constant. It's obvious the drives are busy though, the SMART extended tests end up taking ~5 days and my last scrub took 85 hours.
 

Fraoch

Patron
Joined
Aug 14, 2014
Messages
395
Via SSH:

Code:
cat /var/log/messages


Or just click on the messages at the bottom of the web GUI.

I'm not 100% sure that's the syslog, but there's no /var/log/syslog so that would be my guess.
 

Matdif

Explorer
Joined
Oct 10, 2014
Messages
59
After testing every way I can think of it appears that the Marvell controller becomes unstable under load. (Seems to be fine with only the top sata ports being used.) When I use all 6 or more then 3 I get errors from at least half the ports.

Confirmed with ASRock that my bios and firmware were up to date and they sent me a RMA.

Edit: RMA'd the board but other people here have the same problem with multiple boards. The Marvell controller might not be as compatible with FreeBSD as ASRock thinks. Might just get a raid card put in pass through or a sata extension for PCI or something. I really like the board it has everything that is needed except these sata slots are shit.
 
Last edited:
Status
Not open for further replies.
Top