unRAID Server Release 5.0-beta13 Available

[daniel.boone]

My parity drive rebuild completed with no errors. Cksum match 100% for every file rebuilt from parity. The only question now is why did the parity swap/rebuild not work as intend?

[Thornwood]

When you did this was the parity drive still the biggest drive?

 

I think this might be the problem but not sure.

[molesza]

Installed this to two of my servers. First one upgraded from release 5 beta 12a. No problems there!

 

Second machine upgraded from 4.7 and I get this screen when I start up

 

 

Here is a screenshot of the disk status on 4.7

 

 

Seems it is recognising the drives in the correct places so why is it reporting that the disk is incorrect?

[prostuff1]

Installed this to two of my servers. First one upgraded from release 5 beta 12a. No problems there!

 

Second machine upgraded from 4.7 and I get this screen when I start up

 

 

Here is a screenshot of the disk status on 4.7

 

 

Seems it is recognising the drives in the correct places so why is it reporting that the disk is incorrect?

 

The size is different, search in the general support forum for "HPA"

[molesza]

Thanks. I have read up and see that HPA is a backup of bios on one of my disks. What do I need to do to fix this? So can I go into the bios of the machine and disable HPA and then start the array and allow it to rebuild the drive?

[daniel.boone]

Today I had some parity disk errors. This seemed to have happened after I fixed the owners and permissions on about 50% of my arrray. Some of the folders still had root as owners while new folders had nobody. I just chown nobody:users and then chmod 770. Idea was to get things close so I could re-run the utility script. After that I added a couple new service accounts. Soon after that the parity drive red balled. I captured the log file and reboot. I rebooted and now my array thinks my parity drive has been replaced (blue ball).

 

I understand I can tell it to trust the array but since I had errors I don't think that is best. I'm now thinking maybe my power supply can't keep up. It's a corsair 500w with a mix of 7200 and 5400 RPM drives. I've already ordered a 750 watt seasonic since I was planning to migrate to a norco in the future.

syslog-2011-11-02.zip

[Rastarize]

Upgraded and everything was running perfectly, but overnight two errors appeared:

 

Nov  2 14:11:40 Tower kernel: sh[21343]: segfault at b6af9d53 ip b7675e1c sp bfe6c50c error 4 in libc-2.11.1.so[b7635000+15c000] (Errors)

Nov  2 14:11:40 Tower kernel: mdcmd (52): spindown 0 (Routine)

Nov  2 14:11:40 Tower kernel: hdparm[21344]: segfault at b7129693 ip b765be1c sp bff35950 error 4 in libc-2.11.1.so[b761b000+15c000] (Errors)

 

Are these errors worth worrying about (I can post up the full syslog if preferred)?

[burnaby_boy]

Since upgrading my server to 5.0-beta13, I've noticed that there can sometimes be a 20-30 second delay when I'm accessing the status page of the server in my web browser. With 4.7, when I clicked on the bookmark in Firefox I would immediately be taken to the page. Any ideas as to what might be causing the delay? I'm wondering if it might be the Realtek NIC on the motherboard, and have an Intel PRO/1000 GT PCI adapter on order to test. Anyway, I'd appreciate some guidance with this.

Cheers

[madburg]

orange dot = there is data for this share on the cache drive.

 

Cheers, always wondered what that was for!

 

Has anyone tried the Supermicro AOC-USASLP-L8I with b13?  I'm running b12, debating whether to upgrade.  Do you need any special kind of switch for the link aggregation? I'm just running an unmanaged netgear gigabit.

 

If its a cache share only it will be orange as well. Would be nice to see a different color for cache drive only share (yellow, purple?).

[speeding_ant]

Since upgrading my server to 5.0-beta13, I've noticed that there can sometimes be a 20-30 second delay when I'm accessing the status page of the server in my web browser. With 4.7, when I clicked on the bookmark in Firefox I would immediately be taken to the page. Any ideas as to what might be causing the delay? I'm wondering if it might be the Realtek NIC on the motherboard, and have an Intel PRO/1000 GT PCI adapter on order to test. Anyway, I'd appreciate some guidance with this. (Syslog attached)

Cheers

 

Do you notice any drives spinning up?

[burnaby_boy]

No, the last time I logged in, all the drives were spun down.

[jbartlett]

No, the last time I logged in, all the drives were spun down.

I sometimes get that long delay and can hear drives spinning up one by one. When the page displays, I see that the drives are spun down but if I refresh the page, they're all show as spinning.

[jbartlett]

Ever since upgrading to b13, I've been getting one of these in my syslog every hour:

 

Nov  2 23:06:01 NAS crond[1229]: ignoring /var/spool/cron/crontabs/root- (non-existent user)

[aht961]

Ever since upgrading to b13, I've been getting one of these in my syslog every hour:

 

Nov  2 23:06:01 NAS crond[1229]: ignoring /var/spool/cron/crontabs/root- (non-existent user)

 

Please check this

[burnaby_boy]

No, the last time I logged in, all the drives were spun down.

I sometimes get that long delay and can hear drives spinning up one by one. When the page displays, I see that the drives are spun down but if I refresh the page, they're all show as spinning.

I just accessed the server for the first time this morning and the status page came up immediately. All drives were spinning. So does attempting to access the status page in 5.0-beta13 cause certain drives to be spun up, thus causing the delay? This would be different behaviour from 4.7 where the page would load instantly no matter what the status of the drives. I am using a Gigabyte GA-P35-DS3P motherboard (HPA off), ultilizing all 8 onboard SATA connections, a Supermicro AOC-SASLP-MV8 card and an Adaptec 1430SA card.

Cheers

[jbartlett]

I just accessed the server for the first time this morning and the status page came up immediately. All drives were spinning. So does attempting to access the status page in 5.0-beta13 cause certain drives to be spun up, thus causing the delay? This would be different behaviour from 4.7 where the page would load instantly no matter what the status of the drives. I am using a Gigabyte GA-P35-DS3P motherboard (HPA off), ultilizing all 8 onboard SATA connections, a Supermicro AOC-SASLP-MV8 card and an Adaptec 1430SA card.

Cheers

 

It is a different behavior that I've never experienced in 4.7 but have experienced intermediately after taking the dive with beta10.

[jbartlett]

Ever since upgrading to b13, I've been getting one of these in my syslog every hour:

 

Nov  2 23:06:01 NAS crond[1229]: ignoring /var/spool/cron/crontabs/root- (non-existent user)

 

Please check this

 

That page says to report it as a bug. So.... 

[numanuma]

Does simply replacing bzimage and bzroot update the netatalk? I'm guessing it does, but i just wanted to check first.

[dgaschk]

Does simply replacing bzimage and bzroot update the netatalk? I'm guessing it does, but i just wanted to check first.

 

Yes. As stated in the OP.

[burnaby_boy]

I just accessed the server for the first time this morning and the status page came up immediately. All drives were spinning. So does attempting to access the status page in 5.0-beta13 cause certain drives to be spun up, thus causing the delay? This would be different behaviour from 4.7 where the page would load instantly no matter what the status of the drives. I am using a Gigabyte GA-P35-DS3P motherboard (HPA off), ultilizing all 8 onboard SATA connections, a Supermicro AOC-SASLP-MV8 card and an Adaptec 1430SA card.

Cheers

 

It is a different behavior that I've never experienced in 4.7 but have experienced intermediately after taking the dive with beta10.

 

I added an Intel PRO/1000 GT PCI Ethernet adapter card to my server and disabled the onboard Realtek chip. What a difference! - the delay in accessing the unRaid status page is gone and the transfer times - both ways - seem to have improved.

[madburg]

Realtek's certificate has been stolen and being used for malicious activity at the very least ("Duqu")

[burnaby_boy]

I added an Intel PRO/1000 GT PCI Ethernet adapter card to my server and disabled the onboard Realtek chip. What a difference! - the delay in accessing the unRaid status page is gone and the transfer times - both ways - seem to have improved.

 

Spoke too soon - I tried to access the server this evening and there was a delay again. Initially most of the drives were shown as being spun down. I then refreshed the page and then it showed most of the drives as spinning - very peculiar.

Cheers

[generalz]

I added an Intel PRO/1000 GT PCI Ethernet adapter card to my server and disabled the onboard Realtek chip. What a difference! - the delay in accessing the unRaid status page is gone and the transfer times - both ways - seem to have improved.

 

Spoke too soon - I tried to access the server this evening and there was a delay again. Initially most of the drives were shown as being spun down. I then refreshed the page and then it showed most of the drives as spinning - very peculiar. (syslog attached)

Cheers

 

Mine does it to. despite what people are saying with certain cards it seems to be waking the drives up to get the temp or something when accessing the main page.

[dgaschk]

I added an Intel PRO/1000 GT PCI Ethernet adapter card to my server and disabled the onboard Realtek chip. What a difference! - the delay in accessing the unRaid status page is gone and the transfer times - both ways - seem to have improved.

 

Spoke too soon - I tried to access the server this evening and there was a delay again. Initially most of the drives were shown as being spun down. I then refreshed the page and then it showed most of the drives as spinning - very peculiar. (syslog attached)

Cheers

 

Mine does it to. despite what people are saying with certain cards it seems to be waking the drives up to get the temp or something when accessing the main page.

 

Are you using any add-ons?

[jbartlett]

The spin-ups on access happened to me with a stock beta installation.