September 25, 200619 yr Okay for starters here's what occured this morning as posted to another thread.. Is beta 2 FSCK on reboot? I had a lockup with Beta 2 this morning and had to power cycle - first lockup. Upon bootup it was taking FOREVER to become available again. Panicknig a bit I put a monitor on it and saw filenames for my MP3 collection flying past with messages stating that the record already existed. Swallowing hard I waited for that to end and saw it do several other drives before becoming available. It SEEMs fine now, was it doing a FSCK? What was up as this is the first I've seen this and it was a little scary because it wasn't available at all until this was done.. When I left at o'dark thirty this morning a parity check was ongoiong. No biggie, these are usually pretty fast and while it started out slow this usually speeds up over time. Well, when I got home tonight some 9+ hours later I found that the parity check is STILL ongoing and that it is ONLY 12% finished. I have had to just now move some data to it and it's speed dropped from 1400K per second to 700K per second. OUCH! I used to be able to do a Parity check in just a few hours, even with Beta 1. This time however it is taking much longer. To date it has found just one error. I'm tailing the log now but from this morning's start till now it's got almost nothing in it. TOP shows CPU usage at 97++% and steady with over 40% of that from unraidd. Mind you in another thread I said my next box would have a 3+Gig CPU, times like this is exactly WHY I'll be doing that <shrug> I see mdrecoveryd using a bunch (fluctuating) and and two smbd sessions using about 15% apiece - likely my file copies to the drive. Memory free is like 7000K FWIW, I have a Gig in it... Tom, any ideas?
September 25, 200619 yr Is this an all-IDE system? How many drives? When you use 'top', what are the top 3 processes shown there?
September 26, 200619 yr BLKMGK, Very strange! I am using the beta2 OS version too. I just did a parity check on my number 2 unRaid server, after replacing my original 300GB parity drive with a new 400GB parity drive, and it took a little less than 6 hours. The update rate started out at about 16MB/sec, got up to around 49MB/sec about 2/3 of the way through and finished around 32MB/sec at the end. Regards, TCIII
September 26, 200619 yr The parity check/sync rate is displayed as a "rolling average" of the last 30-second time period. It's showing the rate at which the parity drive is being written. Though the rate might seem low, the amount of data moving through the sytsem is the rate times the number of drives being accessed. The bottleneck is the PCI buses. The rate can flunctuate for several reasons: - If normal i/o is occurring, preference is given to it, so the check/sync rate could decrease. - At the beginning, all the disks are active. If you have a mix of drive sizes, then as check/sync progresses beyond their sizes, there are fewer disks being read, and the rate will increase. When only the parity disk remains, sync simply starts writing zeros and check simply compares that data is all zeros. - On the otherhand, as parity check/sync progresses, the actual data rate of the disks decreases considerably. For most disks up to 50% slower. These factors do not account for BLKMGK's parity check rate decrease - something else is not right.
September 26, 200619 yr I saw terrible speeds when I had a bad cable, a bad moblie rack... just a thought!!! Also, I ran parity last night replacing my 400GB ide with 500GB sata and it was moving very fast!
September 26, 200619 yr Author It's now 36% done at 1400K per second. This is as slow as ever TOP output to be pasted below. This is an ALL IDE system with a total of 12 drives - Intel MB, Gig of RAM, data isn't being written to the drive except in bursts when I copy over a ripped DVD which hasn't been often. Not sure what is up but I do see a single Zombie process and am VERY tempted to recycle. I have found 9 parity errors so far, fingers crossed that a FSCK isn't required I have done parity checks before with this many drives and not been this slow, glad to hear it's just me though and not the code! [size=10pt]Tasks: 41 total, 4 running, 36 sleeping, 0 stopped, 1 zombie Cpu(s): 0.0% user, 99.4% system, 0.0% nice, 0.6% idle Mem: 904924k total, 897972k used, 6952k free, 114140k buffers Swap: 0k total, 0k used, 0k free, 711140k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 983 root 20 0 0 0 0 S 50.6 0.0 1047:43 unraidd 960 root 12 0 0 0 0 R 23.2 0.0 154:33.52 mdrecoveryd 6 root 9 0 0 0 0 S 2.1 0.0 23:17.16 kupdated 2036 root 9 0 1052 1052 840 R 0.6 0.1 0:00.49 top 1 root 9 0 240 240 212 S 0.0 0.0 1:24.09 init 2 root 8 0 0 0 0 S 0.0 0.0 0:00.00 keventd 3 root 19 19 0 0 0 R 0.0 0.0 2:56.74 ksoftirqd_CPU0 4 root 9 0 0 0 0 S 0.0 0.0 2:32.82 kswapd 5 root 9 0 0 0 0 S 0.0 0.0 16:23.59 bdflush 9 root 9 0 0 0 0 S 0.0 0.0 0:00.00 khubd 15 root 9 0 0 0 0 S 0.0 0.0 0:00.00 usb-storage-0 16 root 9 0 0 0 0 S 0.0 0.0 0:00.00 scsi_eh_0 65 root 9 0 680 676 572 S 0.0 0.1 0:12.67 syslogd 68 root 9 0 456 456 400 S 0.0 0.1 0:00.00 klogd 230 root 9 0 0 0 0 S 0.0 0.0 0:00.00 scsi_eh_1 231 root 9 0 0 0 0 S 0.0 0.0 0:00.00 scsi_eh_2 896 root 9 0 516 516 456 S 0.0 0.1 0:00.02 inetd 902 root 9 0 624 624 536 S 0.0 0.1 0:12.22 crond 905 root 9 0 552 552 492 S 0.0 0.1 0:00.00 acpid 917 root 8 0 1152 1152 880 S 0.0 0.1 0:00.87 emhttp 925 root 9 0 520 520 464 S 0.0 0.1 5:11.41 ifplugd 934 root 9 0 0 0 0 Z 0.0 0.0 0:00.00 mdrecoveryd <defunct> 954 root 9 0 1364 1364 1088 S 0.0 0.2 0:00.26 bash 955 root 9 0 484 484 432 S 0.0 0.1 0:00.00 agetty 956 root 9 0 484 484 432 S 0.0 0.1 0:00.00 agetty 957 root 9 0 484 484 432 S 0.0 0.1 0:00.00 agetty 958 root 9 0 484 484 432 S 0.0 0.1 0:00.00 agetty 959 root 9 0 484 484 432 S 0.0 0.1 0:00.00 agetty 980 root 9 0 1152 1152 880 S 0.0 0.1 1:40.97 emhttp 1018 root 9 0 1152 1152 880 S 0.0 0.1 2:13.91 emhttp 1019 root 9 0 0 0 0 S 0.0 0.0 0:37.69 kreiserfsd 1048 root 9 0 1376 1376 1124 S 0.0 0.2 4:19.23 nmbd 1050 root 8 0 2116 2116 1976 S 0.0 0.2 0:01.76 smbd 1051 root 9 0 2104 2104 1964 S 0.0 0.2 0:00.00 smbd 1052 root 9 0 3672 3672 2664 S 0.0 0.4 38:58.37 smbd 1423 root 9 0 824 820 664 S 0.0 0.1 0:00.26 in.telnetd 1425 root 9 0 1356 1356 1088 S 0.0 0.1 0:00.49 bash 1437 root 9 0 632 632 544 S 0.0 0.1 3:38.01 tail 1473 root 9 0 832 828 664 R 0.0 0.1 0:00.60 in.telnetd 1475 root 9 0 1356 1356 1088 S 0.0 0.1 0:00.20 bash 1502 root 9 0 2876 2876 2472 S 0.0 0.3 0:13.27 smbd[/size]
September 26, 200619 yr Next to try is this: hdparm -i /dev/hdha hdparm -i /dev/hdhb hdparm -i /dev/hdhc hdparm -i /dev/hdhd hdparm -i /dev/hdhe hdparm -i /dev/hdhf hdparm -i /dev/hdhg hdparm -i /dev/hdhh hdparm -i /dev/hdhi hdparm -i /dev/hdhj hdparm -i /dev/hdhk hdparm -i /dev/hdhl Each of these commands causes the low-level details of each hard disk to be printed out which looks like this: /dev/hda: Model=ST3500841A, FwRev=3.AAD, SerialNo=3PM03QFX Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode What I'm interested in is the 3 lines which start "PIO modes", "DMA modes", and "UDMA modes". There will be an asterisk next to one of these (in above example it's next to udma5). This will tell us if one of your drives is not in proper DMA mode, which would account for your slowdown (assuming syslog isn't filling up with error messages, which you haven't indicated).
September 26, 200619 yr Author /dev/hdha: No such file or directory is what I got but I think this is a typo, how does hdparm -i /dev/hda sound? Lemme' know if you want me to run other things. I have ALOT of data being transferred to the system right now as I'm ripping DVDs. It is taking FOREVER to get anything onto it and my Parity check is down to 800K per second as a result. I'm tempted tocancel the parity in hopes of just getting the data onto the drive. Prior to this problem speed was noticably faster so yeah I'm a bit puzzled <shrug> [size=10pt]hda: Model=ST3400632A, FwRev=3.04, SerialNo=5NF108P4 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=16384kB, MaxMultSect=16, MultSect=off CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdb: Model=ST3400632A, FwRev=3.04, SerialNo=5NF0Z9KP Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=16384kB, MaxMultSect=16, MultSect=off CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdc: Model=ST3400620A, FwRev=3.AAD, SerialNo=3QG043W1 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=16384kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdd: Model=ST3400620A, FwRev=3.AAC, SerialNo=5QH00QT6 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=16384kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hde: Model=ST3300831A, FwRev=3.03, SerialNo=4NF08RMG Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdf: Model=WDC WD3000JB-00KFA0, FwRev=08.05J08, SerialNo=WD-WCAMR1641443 Config={ HardSect NotMFM HdSw>15uSec SpinMotCtl Fixed DTR>5Mbs FmtGapReq } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=65 BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdg: Model=WDC WD3000JB-00KFA0, FwRev=08.05J08, SerialNo=WD-WCAMR1587689 Config={ HardSect NotMFM HdSw>15uSec SpinMotCtl Fixed DTR>5Mbs FmtGapReq } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=65 BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdh: Model=WDC WD3200JB-22KFA0, FwRev=08.05J08, SerialNo=WD-WCAMR1925507 Config={ HardSect NotMFM HdSw>15uSec SpinMotCtl Fixed DTR>5Mbs FmtGapReq } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=65 BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdi: Model=ST3300831A, FwRev=3.03, SerialNo=4NF08RJX Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdj: Model=ST3300831A, FwRev=3.03, SerialNo=3NF0WH07 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdk: Model=ST3300831A, FwRev=3.01, SerialNo=3NF06PTA Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: * signifies the current active mode [hr] hdl: Model=Maxtor 6L300R0, FwRev=BAJ41G20, SerialNo=L60MYRFG Config={ Fixed } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57 BuffType=DualPortCache, BuffSize=16384kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 udma6 AdvancedPM=yes: disabled (255) WriteCache=enabled Drive conforms to: (null): * signifies the current active mode[/size]
September 26, 200619 yr I wonder if it is something with Seagate drives? I just sent two back because they kept killing my system. I was able to get through 1 of them on SeaTools, but the other would keep hanging. I formated it in XP and it seemed fine and then finally passed the SeaTools.. I said screw it, and I am going to warranty and see what happens
September 26, 200619 yr Author Been using Seagate drives since day one - 5yr warranty and they haven't been failing on me. The system has run parity about 3-4 times with these drives as I installed bigger ones and as I brought new software online. Never have I had an issue. I seriously think that a reboot might solve this but I'm not touching it in case Tom wants to troubleshoot something. I have about 5DVDs worth of data slowly transferring too so I'd like to get those copied before recycling anyway. I'm no longer adding jobs to the que though as it's going to take forever to get these loaded. Syslog had this in it -> root@BEAST:~# tail -f /var/log/syslogtail -f /var/log/syslog tail: cannot open `/var/log/syslogtail' for reading: No such file or directory ==> /var/log/syslog <== [size=10pt]Sep 25 22:11:30 BEAST smbd[1053]: [2006/09/25 22:11:30, 0] lib/util_sock.c:write _socket(455) Sep 25 22:11:30 BEAST smbd[1053]: write_socket: Error writing 4 bytes to socke t 22: ERRNO = Connection reset by peer Sep 25 22:11:30 BEAST smbd[1053]: [2006/09/25 22:11:30, 0] lib/util_sock.c:send_ smb(647) Sep 25 22:11:30 BEAST smbd[1053]: Error writing 4 bytes to client. -1. (Connec tion reset by peer) Sep 25 22:16:44 BEAST nmbd[1048]: [2006/09/25 22:16:44, 0] nmbd/nmbd_become_lmb. c:become_local_master_stage2(396) Sep 25 22:16:44 BEAST nmbd[1048]: ***** Sep 25 22:16:44 BEAST nmbd[1048]: Sep 25 22:16:44 BEAST nmbd[1048]: Samba name server BEAST is now a local maste r browser for workgroup PEARL on subnet 192.168.1.101 Sep 25 22:16:44 BEAST nmbd[1048]: Sep 25 22:16:44 BEAST nmbd[1048]: ***** Sep 26 19:50:09 BEAST smbd[1742]: [2006/09/26 19:50:09, 0] lib/util_sock.c:get_peer_addr(1000) Sep 26 19:50:09 BEAST smbd[1742]: getpeername failed. Error was Transport endpoint is not connected Sep 26 19:50:10 BEAST smbd[1742]: [2006/09/26 19:50:10, 0] lib/util_sock.c:get_peer_addr(1000) Sep 26 19:50:10 BEAST smbd[1742]: getpeername failed. Error was Transport endpoint is not connected Sep 26 19:50:10 BEAST smbd[1742]: [2006/09/26 19:50:10, 0] lib/util_sock.c:write_socket_data(430) Sep 26 19:50:10 BEAST smbd[1742]: write_socket_data: write failure. Error = Connection reset by peer Sep 26 19:50:10 BEAST smbd[1742]: [2006/09/26 19:50:10, 0] lib/util_sock.c:write_socket(455) Sep 26 19:50:10 BEAST smbd[1742]: write_socket: Error writing 4 bytes to socket 5: ERRNO = Connection reset by peer Sep 26 19:50:10 BEAST smbd[1742]: [2006/09/26 19:50:10, 0] lib/util_sock.c:send_smb(647) Sep 26 19:50:10 BEAST smbd[1742]: Error writing 4 bytes to client. -1. (Connection reset by peer) Sep 26 20:12:34 BEAST smbd[1460]: [2006/09/26 20:12:34, 0] lib/util_sock.c:read_socket_data(384) Sep 26 20:12:34 BEAST smbd[1460]: read_socket_data: recv failure for 4. Error = Connection timed out Sep 27 06:09:09 BEAST kernel: hda: irq timeout: status=0x80 { Busy } Sep 27 06:09:09 BEAST kernel: Sep 27 06:09:15 BEAST kernel: ide0: reset: success Sep 27 06:09:15 BEAST kernel: md0: parity incorrect: 281613328 Sep 27 06:09:15 BEAST kernel: md0: parity incorrect: 281613336 Sep 27 06:09:15 BEAST kernel: md0: parity incorrect: 281613344 Sep 27 06:09:15 BEAST kernel: md0: parity incorrect: 281613352 Sep 27 06:09:15 BEAST kernel: md0: parity incorrect: 281613360 Sep 27 06:09:15 BEAST kernel: md0: parity incorrect: 281613368 Sep 27 06:09:15 BEAST kernel: md0: parity incorrect: 281613376 Sep 27 06:09:15 BEAST kernel: md0: parity incorrect: 281613384[/size] Nothing too exciting IMO. Tom was that a FSCK I saw when the system first fired up or soemthing else? I've never seen those sorts of messages roll by before....
September 27, 200619 yr Author Okay, I stopped all of the file transfers. The network usage was spiking up to as high as 1% so it was pretty apparent not much data was moving. Parity speed is up to a whopping 800K Some additional data in my Syslog after I did this... [size=10pt]Sep 27 06:09:15 BEAST kernel: md0: parity incorrect: 281613384 Sep 27 10:56:21 BEAST smbd[2329]: [2006/09/27 10:56:21, 0] smbd/oplock.c:request_oplock_break(1055) Sep 27 10:56:21 BEAST smbd[2329]: request_oplock_break: no response received to oplock break request to pid 1052 on port 32769 for dev = 908, inode = 63102, file_id = 2930 Sep 27 10:57:38 BEAST smbd[2329]: [2006/09/27 10:57:38, 0] smbd/oplock.c:request_oplock_break(1055) Sep 27 10:57:38 BEAST smbd[2329]: request_oplock_break: no response received to oplock break request to pid 1052 on port 32769 for dev = 908, inode = 63104, file_id = 2940 Sep 27 10:57:50 BEAST smbd[2411]: [2006/09/27 10:57:50, 0] smbd/oplock.c:request_oplock_break(1055) Sep 27 10:57:50 BEAST smbd[2411]: request_oplock_break: no response received to oplock break request to pid 1052 on port 32769 for dev = 908, inode = 37745, file_id = 4506 Sep 27 10:57:50 BEAST smbd[2411]: [2006/09/27 10:57:50, 0] smbd/open.c:open_mode_check(734) Sep 27 10:57:50 BEAST smbd[2411]: open_mode_check: exlusive oplock left by process 1052 after break ! For file Shows/Dr. Who 2005/Season 2/Doctor.Who.2005.S02E11.WS.PDTV.XviD-FoV.[TvT].avi, dev = 908, inode = 37745. Deleting it to continue... Sep 27 10:57:50 BEAST smbd[2411]: [2006/09/27 10:57:50, 0] smbd/open.c:open_mode_check(738) Sep 27 10:57:50 BEAST smbd[2411]: open_mode_check: Existent process 1052 left active oplock. Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] smbd/oplock.c:oplock_break(830) Sep 27 10:58:03 BEAST smbd[1052]: oplock_break: end of file from client Sep 27 10:58:03 BEAST smbd[1052]: oplock_break failed for file DVD/RUNAWAY_JURY/VIDEO_TS/VTS_04_1.VOB (dev = 908, inode = 63102, file_id = 2930). Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] smbd/oplock.c:oplock_break(923) Sep 27 10:58:03 BEAST smbd[1052]: oplock_break: client failure in break - shutting down this smbd. Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] tdb/tdbutil.c:tdb_log(725) Sep 27 10:58:03 BEAST smbd[1052]: tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 127 ltype=0 (Resource deadlock avoided) Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] tdb/tdbutil.c:tdb_log(725) Sep 27 10:58:03 BEAST smbd[1052]: tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 59 ltype=1 (Resource deadlock avoided) Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] tdb/tdbutil.c:tdb_log(725) Sep 27 10:58:03 BEAST smbd[1052]: tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 59 ltype=0 (Resource deadlock avoided) Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] tdb/tdbutil.c:tdb_log(725) Sep 27 10:58:03 BEAST smbd[1052]: tdb(/var/cache/samba/locking.tdb): tdb_unlock: count is 0 Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] tdb/tdbutil.c:tdb_log(725) Sep 27 10:58:03 BEAST smbd[1052]: tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 127 ltype=1 (Resource deadlock avoided) Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] tdb/tdbutil.c:tdb_log(725) Sep 27 10:58:03 BEAST smbd[1052]: tdb(/var/cache/samba/locking.tdb): tdb_lock failed on list 127 ltype=0 (Resource deadlock avoided) Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] tdb/tdbutil.c:tdb_log(725) Sep 27 10:58:03 BEAST smbd[1052]: tdb(/var/cache/samba/locking.tdb): tdb_unlock: count is 0 Sep 27 10:58:03 BEAST smbd[1052]: [2006/09/27 10:58:03, 0] smbd/oplock_linux.c:linux_oplock_receive_message(143) Sep 27 10:58:03 BEAST smbd[1052]: Invalid file descriptor 25 in kernel oplock break![/size]
September 27, 200619 yr BLKMGK, I'm seeing this on my beta2 box as well. A few days ago I rebuilt my server adding 2 400G Seagate drives. Everything went smooth as silk. Today, I get home from work and see that my parity disk had been stopped due to "errors". I rebooted the array (probably not the wisest thing to do) and now I'm rocketting through parity check at about 11 KB/sec. :'( This is an all ide box, recommended hardware, 512MB ram, and 5 drives in the array. Any chance you could post some 'baby steps' for a Linux noob on how you troubleshoot this. I see referance to "TOP", and logging, but have no idea how to do either. Thanks! -PGPfan
September 27, 200619 yr Author To tail the syslog and lookk for system errors try this -> tail -f /var/log/syslogtail -f /var/log/syslog in a telnet window. Top is just that - in a Telnet window type TOP and enter. To get out hit CTRL-C. It's like Task Manager and yeah you can kill threads this way but I wouldn't recommend it Guess my three new drives will be awhile being installed (lol)
September 27, 200619 yr I've also run into a very slow parity. I'm averaging 1,590 KB/sec, the estimated finish is in 4050 minutes! I tried running the hdparm command (on hadha/hda.. etc) and get an error- "no such file or directory"... Do I need to stop party to try this command? Equipment- nearly all standard fare, recommended Intel MB, recommended Promise SATA 300 tx4s) three identical WD WD4000YR drives, just formated, gig of ram, and Cheiftec SATA2131B mobilerack (all items previously tested in 3ware hw raid). -Alex
September 27, 200619 yr Author Look above at the EXACT command I ran for HDPARM, tom had a typo in his post to me...
September 27, 200619 yr I've also run into a very slow parity. I'm averaging 1,590 KB/sec, the estimated finish is in 4050 minutes! I tried running the hdparm command (on hadha/hda.. etc) and get an error- "no such file or directory"... Do I need to stop party to try this command? Equipment- nearly all standard fare, recommended Intel MB, recommended Promise SATA 300 tx4s) three identical WD WD4000YR drives, just formated, gig of ram, and Cheiftec SATA2131B mobilerack (all items previously tested in 3ware hw raid). -Alex What brand drives do you have in your system?
September 27, 200619 yr Author I have mostly Seagate now, one WD I think, some Maxtor. I do not think it's th edrives since it ran fine several times previously and has only just now slowed on this new relrease after having worked okay before. Priot to the lockup and cold boot it was running fine, now slow as a turtle even for file transfers it seems like.
September 27, 200619 yr I have mostly Seagate now, one WD I think, some Maxtor. I do not think it's th edrives since it ran fine several times previously and has only just now slowed on this new relrease after having worked okay before. Priot to the lockup and cold boot it was running fine, now slow as a turtle even for file transfers it seems like. Right, I hear you.... I was just wondering because the moment I removed the 2 Seagate drives from my array things went back to normal.... I don't know, I just don't understand because I am going to RMA these drives, but I was able to run SeaTools on them, but the moment I put them into the array things were very slow like you are saying... Just trying to help
September 27, 200619 yr I run all identical Western Digital 400gb raid drives (model WD4000YR). It's been 24 hours and I'm only 17% done with the parity on 2 empty/formated drives Here is the configuration: parity device: /dev/hdd (hdd1) PATA:WDC WD4000YR-01PLB0/WD-WMAMY1204302 disk1 device: /dev/hdc (hdc1) PATA:WDC WD4000YR-01PLB0/WD-WMAMY1215357 disk2 device: /dev/scsi/sdh1-1c0i0l0 (sdb1) SATA:WDC WD4000YR-01P/WD-WMAMY1215702 Oddly it lists the drives connected to the Motherboard as "PATA" whereas the same type of drive off of the Promise card as SATA, which it is, that couldnt be the issue could it?
September 27, 200619 yr I run all identical Western Digital 400gb raid drives (model WD4000YR). It's been 24 hours and I'm only 17% done with the parity on 2 empty/formated drives Here is the configuration: parity device: /dev/hdd (hdd1) PATA:WDC WD4000YR-01PLB0/WD-WMAMY1204302 disk1 device: /dev/hdc (hdc1) PATA:WDC WD4000YR-01PLB0/WD-WMAMY1215357 disk2 device: /dev/scsi/sdh1-1c0i0l0 (sdb1) SATA:WDC WD4000YR-01P/WD-WMAMY1215702 Oddly it lists the drives connected to the Motherboard as "PATA" whereas the same type of drive off of the Promise card as SATA, which it is, that couldnt be the issue could it? What motherboard? It sounds like the motherboard SATA controllers are set up in "compatibility mode", where it makes SATA drives look like IDE (aka, PATA) drives. Try going into your bios and look for "enhanced" mode or "native SATA" mode.
September 27, 200619 yr I thought I did put it in enchanced mode, I'll check again! Motherboard is the Intel D865GLC.
September 27, 200619 yr Author Tom any ideas in my case? Do you need more information on my setup? Should I drop back to Beta1 as a test?
September 27, 200619 yr Tom any ideas in my case? Do you need more information on my setup? Should I drop back to Beta1 as a test? Actually, we're going to post a beta4 soon which might help with the lock-up issue. The troublesome entry in your syslog is: Sep 27 06:09:09 BEAST kernel: hda: irq timeout: status=0x80 { Busy } Sep 27 06:09:09 BEAST kernel: Sep 27 06:09:15 BEAST kernel: ide0: reset: success I'm thinking that even though all your drives claim to be in udma mode, one of the IDE interfaces is using PIO mode. In beta4 we went back to the pre-3.0 method of interrupt handling which might fix this problem. Regarding your observation of seeing the names of music files flying by during bootup - this is not an fsck. These are messages being output that indicate duplicate file names in a 'composite' directory. For example, suppose you have this: disk1/Music/song1.mp3 disk1/Music/song2.mp3 disk2/Music/song2.mp3 disk2/Music/song3.mp3 Well song2.mp3 is in both disk1/Music and disk2/Music. When the User Shares are built, the one in disk1 will show up in the Music user share, and the one in disk2 will not show up at all, but system will print a message saying duplicate file found. Do you see this possibly happening with your directory structure?
September 27, 200619 yr Actually, we're going to post a beta4 soon which might help with the lock-up issue. The troublesome entry in your syslog is: In case anyone is wondering what happend to beta3 - that was a special build for a single customer that added VIA sata support.
September 27, 200619 yr I'm the guilty party for beta3 - and it ended up there is a "limitation" with the Via chip as well so I ended up ditching it. Just waiting for the new Asus board to arrive. Anyway, never let it be said that Tom won't go out of his way to help somebody out! ...and now back to normal thread transmission....
Archived
This topic is now archived and is closed to further replies.