March 6, 201313 yr I read and re-read the UnRaid troubleshooting. I have not found a solution that works yet. Problem begins here. I have one with 22 drives and it works like a champ. I say this as I have added drives several times in the past. Server 1 Lian Li Full tower with 1250W power supply and 4@5-in-3 supermicro cages with Noctua fans. 16 Gbytes memory 2 Supermicro HBA cards ARock A75 Extreme6 motherboard with 3 PCIe 16x slots and 8 SATA3 ports on Motherboard. Cache and parity are Hitachi 4TB Ultrastar Total Data Drive capacity 45TB I am rebuilding my second Unraid and I am having a few issues. I moved a smaller UnRaid system into a RPC-4224 case with 1 parity and 5 data drives. 120mm Fan Wall with 2@80mm Noctu exhaust fans 8 Gbytes memory 2 Supermicro HBA cards Only one HBA is installed at the moment. ARock A75 Extreme6 motherboard with 3 PCIe 16x slots and 8 SATA3 ports on Motherboard. I took all my extra drives and was going to add them into the unraid. I precleared 8 Seagate Barricuda 3TB drives using the pre-clear script and Screen. I powered up the UnRaid system and it came up with zero difficulty. All my drives report as SATA (sdx). All 6 existing drives are on the Motherboard SATA3 ports. I plugged a new Seagate drive into the 7th Motherboard SATA3 port and it was detected correctly by the system. I stopped the array and went to add the new drive. The add drive process started, and then I lost the WEBGUI from the system. I could no longer login to the system using "//tower" I could still see something was happening from UnMenu. It appeared that something was writing to the new drive. The counter kept incrementing I waited for over 8 hours, but the system never recovered. Upon rebooting the system, I was able to use the WebGui again. I tried several different drives one at time in the same and different SATA ports/cables. Any ideas what is happening? I thank you in advance for trying to help me. Syslog is attached syslog-2013-03-05.txt
March 6, 201313 yr You've got some "bad method" unmenu errors that I've seen when two instances of unmenu are running, but other than cluttering up the syslog they are harmless. It appears the disk you were adding was not precleared and there were some cable/controller/drive errors where it looks like a drive dropped offline and then came back. I'd check the cabling. Here is the interesting part of the log: Mar 5 11:59:33 Tower emhttp: clearing disk8... .... Mar 5 12:46:52 Tower kernel: drivers/scsi/mvsas/mv_sas.c 2094:Get signature time out, reset phy 2 .... Mar 5 12:46:54 Tower kernel: mvsas 0000:01:00.0: Phy2 : No sig fis Mar 5 12:46:54 Tower kernel: drivers/scsi/mvsas/mv_sas.c 2052:phy2 Attached Device Mar 5 12:46:55 Tower kernel: drivers/scsi/mvsas/mv_sas.c 2108:phy 1 ctrl sts=0x00199800. Mar 5 12:46:55 Tower kernel: drivers/scsi/mvsas/mv_sas.c 2110:phy 1 irq sts = 0x00010000 Mar 5 12:46:55 Tower kernel: drivers/scsi/mvsas/mv_sas.c 2159:notify plug in on phy[1] Mar 5 12:46:55 Tower kernel: drivers/scsi/mvsas/mv_sas.c 1205:phy 1 attach dev info is 0 Mar 5 12:46:55 Tower kernel: drivers/scsi/mvsas/mv_sas.c 1207:phy 1 attach sas addr is 1 Mar 5 12:46:55 Tower kernel: drivers/scsi/mvsas/mv_sas.c 271:phy 1 byte dmaded. Mar 5 12:46:55 Tower kernel: sas: phy-1:1 added to port-1:1, phy_mask:0x2 ( 100000000000000) Mar 5 12:46:55 Tower kernel: sas: DOING DISCOVERY on port 1, pid:922 Mar 5 12:46:55 Tower kernel: drivers/scsi/mvsas/mv_sas.c 1371:found dev[1:5] is gone. Mar 5 12:46:55 Tower kernel: sas: Enter sas_scsi_recover_host Mar 5 12:46:55 Tower kernel: ata9: sas eh calling libata port error handler Mar 5 12:46:55 Tower kernel: ata10: sas eh calling libata port error handler Mar 5 12:46:55 Tower kernel: sas: sas_ata_hard_reset: Found ATA device. Mar 5 12:46:55 Tower kernel: ata10.00: ATA-8: ST3000DM001-9YN166, CC4B, max UDMA/133 Mar 5 12:46:55 Tower kernel: ata10.00: 5860533168 sectors, multi 0: LBA48 NCQ (depth 31/32) Mar 5 12:46:55 Tower kernel: ata10.00: configured for UDMA/133 Mar 5 12:46:55 Tower kernel: sas: --- Exit sas_scsi_recover_host Mar 5 12:46:55 Tower kernel: scsi 1:0:1:0: Direct-Access ATA ST3000DM001-9YN1 CC4B PQ: 0 ANSI: 5 Mar 5 12:46:55 Tower kernel: sas: DONE DISCOVERY on port 1, pid:922, result:0
March 6, 201313 yr Author Thank You for looking at it. yes, I know how that disconnected drive happened. after waiting for hours for the system to add the drives i tried reseating the new drives. All my SATA/8087 cables are the "locking" variety. I rechecked them all anyway and they all worked and were tight. since this is a "new" array (not much data and ALL data fully backed up) i have decided to go with the "initconfig" option to force the addition of the new drives. First, I am running preclear again on all new drives. The drives that showed as NOT precleared, had actually been precleared. the only activity was the drive adding process. not quite sure how i might have gotten 2 instances of unmenu running. I will check running processes. after using this "nuclear" "initconfig" option I will report back on whether or not I need to start building the array from scratch. Gary
March 6, 201313 yr not quite sure how i might have gotten 2 instances of unmenu running. I will check running processes. The "Bad Method" messages are emitted from unMENU when it is probed by something on your LAN on port 8080 that is not a "GET" or "POST" operation. These have typically been some other PC or device attached to your LAN attempting to locate open ports and shares and auto-configure. The messages are not caused by two versions of unMENU running, as that is not even possible, as only one would be able to open up the port 8080 for communications. The other person suggesting that as the cause is misinformed.
Archived
This topic is now archived and is closed to further replies.