jayhawk

I'm at a full slack build with b10 - and one sas supermicro card, it's been solid. if i add another sas card, i quickly start to see errors. I could swap either card back and forth and its fine, using one. So, not sure if that helps but that's what I've seen. I bought some adaptec 4 port sata cards to use along with the onboard and sas card.

I hate to ring the, "hey can we have an update bell"--but for those of us actively testing this we seem to be sliding off support mountain. Yes, I realize this is a beta, however, updates were historically coming fast enough that if there were a bug or two--one could tolerate the bug, work through it, wait patiently and continue testing. We're pushing a month without a new beta revision, and weeks without a word from LimeTech (last post 15 days ago forum wide). A little discouraging for some that are testing with hardware that isn't currently working properly, and want to continue to help. flame on.

Not that anything is 'owed' to anyone that is running beta--but an update or eta for next revision 12b, 13, whatever it may be, would be nice. As the last poster(s) indicated, it is beta--so, whatever changes have been made in an attempt to stabilize the BLK issue, network driver issues, I'm sure everyone would be willing to continue to test and report whatever progress has been made in the last couple weeks. For me, I reduced my server to 1 supermicro sas card, added 2 adaptec cards, moved to a different case (avoiding the norco 4224 backplane) and tried again. With two supermicro sas cards in, I still saw the issues. When I removed it, and added the adaptec card(s) and used onboard--I've yet to see that lockup issue. I don't know if it's an issue with the supermicro board (motherboard/pcie) or the sas cards, or the drivers/unraid... but seems it likes one card a lot better than multiple. For now it's stable for me. I've decided to try building it into a full release of slack again--I really need a full linux box and without the ability to do the CPIO/zcat mod to permanently add packages, kindof a must for me.

Mainly, its a hassle getting everything working as I need/want it. I'd like to be able to load whatever I need to run a standalone system (so slack may be a better choice in the end)--without having to wait for 20-40 pkg's to install at run time. I also haven't seen an apache plugin--which I use for mod_proxy/rewriting all the different ports to URL's--easier to remember. That's another thing that unRAID/unWEB/unMENU are generally not friendly with since you can't set a web/context root. Overall I think it's just the ease of configuring it like a standard linux box--the plugins just get confusing, upgrading things is more confusing. If i can just load what I need and not have to worry about the thing resetting any changes I made because it loaded to the memory and not actual disk... that's what I want/need. I'll dig around for the slack build with unraid 5x--It would have been nice if this would still work, or someone could chip in a root cause for why it wont. I'm more willing to accept it if it has a reason as to why, functionally.

So only choice at this point is plugins? anyone running 5betaX successfully with a full build of slack 13.37?

I'll have another look for the samba/afp exclusives--it would be nice to enable show/hide or so it only see's one device. instead of: tower tower-SMB etc. Thanks for the response on the disk export, that was killing me

I've played around with the share settings in the new 5 series. I can't seem to not share out all the disks. If i restrict which disks Im sharing (excluded) it starts excluding any shares with those disks in them as well. Any idea how thats supposed to work, or if there's something im missing? Further to that, I'd like to be able to share things through just NFS, just SAMBA, or just AFP... i.e.: I want to share backup via AFP I want to share movies via SAMBA but i dont want the backup to be share via SAMBA or the movies share via AFP? or better, can i export SAMBA to just a certain IP range so that only certain machines see the shares, using allow/deny? It's confusing seeing afp/samba shares duplicated in osx (for some).

I was using the instructions I used from 4.7 (and prior iterations) to add in permanent apps. See: http://lime-technology.com/forum/index.php?topic=871.0 Will this technique not work for 5.x betas? I can't seem to get this to work. So far, I've gotten it to boot--but it's not working as it should (no disks, disks show but aren't actually there (size unknown etc). All my applications are loading in, python, httpd, other various support packages... perhaps I'm missing something before I sync? I'd prefer this technique to plugins, but if its not going to work I can fall back... thanks for any insight...

Precleared drives, added them to the array--then when it was idle, apparently the BLK issue came in. It had been up nearly 3 days while it was "busy" preclearing. The first 10 hours it was idle, it dies with the BLK error. Sep 6 08:26:43 HBO in.telnetd[30231]: connect from 10.0.1.1 (10.0.1.1) Sep 6 08:26:48 HBO login[30232]: ROOT LOGIN on '/dev/pts/1' from 'awesome.dog' Sep 6 11:25:01 HBO kernel: sas: command 0xc51f8600, task 0xc307bcc0, timed out: BLK_EH_NOT_HANDLED Sep 6 11:25:01 HBO kernel: sas: Enter sas_scsi_recover_host Sep 6 11:25:01 HBO kernel: sas: trying to find task 0xc307bcc0 Sep 6 11:25:01 HBO kernel: sas: sas_scsi_find_task: aborting task 0xc307bcc0 Sep 6 11:25:01 HBO kernel: drivers/scsi/mvsas/mv_sas.c 1818:<7>mv_abort_task() mvi=f76a0000 task=c307bcc0 slot=f76b1640 slot_idx=x2 Sep 6 11:25:01 HBO kernel: sas: sas_scsi_find_task: querying task 0xc307bcc0 Sep 6 11:25:01 HBO kernel: drivers/scsi/mvsas/mv_sas.c 1747:mvs_query_task:rc= 5 Sep 6 11:25:01 HBO kernel: sas: sas_scsi_find_task: task 0xc307bcc0 failed to abort Sep 6 11:25:01 HBO kernel: sas: task 0xc307bcc0 is not at LU: I_T recover Sep 6 11:25:01 HBO kernel: sas: I_T nexus reset for dev 0600000000000000 Sep 6 11:25:01 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2198:port 6 ctrl sts=0x89800. Sep 6 11:25:01 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2200:Port 6 irq sts = 0x1001 Sep 6 11:25:01 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2226:phy6 Unplug Notice Sep 6 11:25:01 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2198:port 6 ctrl sts=0x199800. Sep 6 11:25:01 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2200:Port 6 irq sts = 0x11081 Sep 6 11:25:01 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2253:notify plug in on phy[6] Sep 6 11:25:01 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2278:plugin interrupt but phy6 is gone Sep 6 11:25:03 HBO kernel: mvsas 0000:03:00.0: Phy6 : No sig fis Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2139:phy6 Attached Device Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2198:port 6 ctrl sts=0x89800. Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2200:Port 6 irq sts = 0x1001 Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2226:phy6 Unplug Notice Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2198:port 6 ctrl sts=0x199800. Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2200:Port 6 irq sts = 0x81 Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2198:port 6 ctrl sts=0x199800. Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2200:Port 6 irq sts = 0x10000 Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2253:notify plug in on phy[6] Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 1338:port 6 attach dev info is 20000 Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 1340:port 6 attach sas addr is 6 Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 379:phy 6 byte dmaded. Sep 6 11:25:03 HBO kernel: sas: sas_form_port: phy6 belongs to port6 already(1)! Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 1701:mvs_I_T_nexus_reset for device[6]:rc= 0 Sep 6 11:25:03 HBO kernel: sas: I_T 0600000000000000 recovered Sep 6 11:25:03 HBO kernel: sas: sas_ata_task_done: SAS error 8d Sep 6 11:25:03 HBO kernel: ata10: sas eh calling libata port error handler Sep 6 11:25:03 HBO kernel: ata11: sas eh calling libata port error handler Sep 6 11:25:03 HBO kernel: ata12: sas eh calling libata port error handler Sep 6 11:25:03 HBO kernel: ata13: sas eh calling libata port error handler Sep 6 11:25:03 HBO kernel: ata14: sas eh calling libata port error handler Sep 6 11:25:03 HBO kernel: ata15: sas eh calling libata port error handler Sep 6 11:25:03 HBO kernel: ata16: sas eh calling libata port error handler Sep 6 11:25:03 HBO kernel: ata16.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 t0 Sep 6 11:25:03 HBO kernel: ata16.00: failed command: CHECK POWER MODE Sep 6 11:25:03 HBO kernel: ata16.00: cmd e5/00:00:00:00:00/00:00:00:00:00/40 tag 0 Sep 6 11:25:03 HBO kernel: res 01/04:ff:00:00:00/00:00:00:00:00/40 Emask 0x3 (HSM violation) Sep 6 11:25:03 HBO kernel: ata16.00: status: { ERR } Sep 6 11:25:03 HBO kernel: ata16.00: error: { ABRT } Sep 6 11:25:03 HBO kernel: ata16: hard resetting link Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2198:port 6 ctrl sts=0x89800. Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2200:Port 6 irq sts = 0x1001 Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2226:phy6 Unplug Notice Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2198:port 6 ctrl sts=0x199800. Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2200:Port 6 irq sts = 0x11081 Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2253:notify plug in on phy[6] Sep 6 11:25:03 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2278:plugin interrupt but phy6 is gone Sep 6 11:25:05 HBO kernel: mvsas 0000:03:00.0: Phy6 : No sig fis Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2139:phy6 Attached Device Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2198:port 6 ctrl sts=0x99800. Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2200:Port 6 irq sts = 0x1081 Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2226:phy6 Unplug Notice Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2198:port 6 ctrl sts=0x199800. Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2200:Port 6 irq sts = 0x10000 Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 2253:notify plug in on phy[6] Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 1338:port 6 attach dev info is 20000 Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 1340:port 6 attach sas addr is 6 Sep 6 11:25:05 HBO kernel: drivers/scsi/mvsas/mv_sas.c 379:phy 6 byte dmaded. Sep 6 11:25:05 HBO kernel: sas: sas_form_port: phy6 belongs to port6 already(1)! Sep 6 11:25:06 HBO kernel: drivers/scsi/mvsas/mv_sas.c 1701:mvs_I_T_nexus_reset for device[6]:rc= 0 Sep 6 11:25:06 HBO kernel: sas: sas_ata_hard_reset: Found ATA device. Sep 6 11:25:06 HBO kernel: ata16.00: configured for UDMA/133 Sep 6 11:25:06 HBO kernel: ata16: EH complete Sep 6 11:25:06 HBO kernel: ata17: sas eh calling libata port error handler Sep 6 11:25:06 HBO kernel: sas: --- Exit sas_scsi_recover_host Sep 6 14:49:22 HBO kernel: sas: command 0xe233a9c0, task 0xd5443a40, timed out: BLK_EH_NOT_HANDLED Sep 6 17:48:02 HBO in.telnetd[23171]: connect from 10.0.1.77 (10.0.1.77) if you need a whole syslog I can provide it, but you can see nothing happening prior to the error.

I know it's a little late in the development stages to request a change/implement new functionality, but it would be really handy if you allowed the ability to set a webroot--so when proxying the ports through apache, the CSS/links worked properly. i.e. normal.unraid.server/Main normal.unraid.server:8080/Main proxyd.unraid.server/unRAID/Main (formatting and redirect problems) The same can be accomplished by not hardcoding urls--but setting webroot if present in front of them would work as well.

I've precleared two drives, and have been streaming movies/tv since upgrading to this version and (knock on wood) have not seen the blk error and subsequent failure. root@HBO:/mnt/# uptime 21:54:13 up 1 day, 3:21, 2 users, load average: 1.86, 1.79, 1.65 Load is from preclear.

hehe, I'll move to 12a and see if the BLK issue goes away. I honestly can't say where/why/when it decides to happen. The box has been up since yesterday afternoon without it appearing in the log. I've also grabbed a 2-ch sata card so I can remove one of the SAS cards and use onboard+2ch. I'll test beta 12a first though, before I start tearing out parts.

I can't be completely certain, but I would definitely start by checking the connections to Disk 4, sdk, the WD20EARS with serial ending in 286. Anyway to tell from the logs, for certain? I'd think that if it were a power issue it would be multiple drives that would have the problem, since its an entire row? If it's a card, I could swapping it / updating firmware--but i just reseated them all last night. I did have a couple cables that were making pretty tight turns/semi-crimped so i straightened them out. After doing all that, it did it within a couple hours of coming up... I restarted it again last night and its been up overnight. I'll wait on the .13 release and wait to see if anyone can tell me definitively which card is showing the ata power up failure. at least then i could move all the drives off of it to see if i have the same problem.

Not trying to beat a dead horse, but can you just ensure that the backplane for the drive that is failing in the Norco is properly connected to both power connectors. I say this because I thought I was fine since all the connectors were connected (main and backup power sockets), but found that the cheap splitter I got actually had a pin that was pushed out of it's crimp, so one of the two connectors wasn't actually providing power... Once I replaced it all was good. I have the same exact setup as you with 3 of those cards in a Norco and have not seen that problem with b12 running for about 4 days straight... Good luck friend. certainly doesnt hurt to check. any way to tell which mvsas controller its complaining about --or disk thats losing power per the logs? perhaps if i knew which connector seemed to be shy of power it would help me get in there and hammer it out.

this happened to me once, i rebooted and it was 'fine' -until i got the BLK_UNHANDLED issue with the SAS controller again think I'll be rolling back to 4.7 today.