pm1961 Posted November 19, 2015 Posted November 19, 2015 Hi, As per the subject title, I keep getting this when I try to write new stuff to my unRAID server. It's a recent change and I don't know what has caused it. I can read from the server normally and all seems to be in order with the system. Is this a windows (8.1) or unRAID issue? During boot, I get down to the "Tower login:" ok and then when I start the Dashboard on a web page the following happens; chmod: changing permissions of '/mnt/user/System Volume Information': Read-only file system This is the last bit of the sys.log Nov 19 16:00:24 Tower logger: mount: warning: /mnt/cache seems to be mounted read-only. Nov 19 16:00:24 Tower kernel: ntfs: volume version 3.1. Nov 19 16:00:24 Tower emhttp: shcmd (60): sync Nov 19 16:00:26 Tower emhttp: shcmd (61): mkdir /mnt/user0 Nov 19 16:00:26 Tower emhttp: shcmd (62): /usr/local/sbin/shfs /mnt/user0 -disks 8190 -o noatime,big_writes,allow_other |& logger Nov 19 16:00:26 Tower emhttp: shcmd (63): mkdir /mnt/user Nov 19 16:00:26 Tower emhttp: shcmd (64): /usr/local/sbin/shfs /mnt/user -disks 8191 2048000000 -o noatime,big_writes,allow_other -o remember=0 |& logger Nov 19 16:00:26 Tower emhttp: shcmd (65): cat - > /boot/config/plugins/dynamix/mover.cron <<< "# Generated mover schedule:#01240 3 * * * /usr/local/sbin/mover |& logger#012" Nov 19 16:00:26 Tower emhttp: shcmd (66): /usr/local/sbin/update_cron &> /dev/null Nov 19 16:00:26 Tower emhttp: shcmd (67): :>/etc/samba/smb-shares.conf Nov 19 16:00:26 Tower emhttp: shcmd (68): chmod 0777 '/mnt/user/System Volume Information' Nov 19 16:00:26 Tower shfs/user: shfs_chmod: chmod: /mnt/cache/System Volume Information 040777 (30) Read-only file system Nov 19 16:00:26 Tower emhttp: shcmd: shcmd (68): exit status: 1 Nov 19 16:00:26 Tower emhttp: Restart SMB... Nov 19 16:00:26 Tower emhttp: shcmd (69): killall -HUP smbd Nov 19 16:00:26 Tower emhttp: shcmd (70): cp /etc/avahi/services/smb.service- /etc/avahi/services/smb.service Nov 19 16:00:26 Tower emhttp: shcmd (71): pidof rpc.mountd &> /dev/null Nov 19 16:00:26 Tower emhttp: shcmd (72): /etc/rc.d/rc.atalk status Nov 19 16:00:26 Tower emhttp: Start AVAHI... Nov 19 16:00:26 Tower emhttp: shcmd (73): /etc/rc.d/rc.avahidaemon start |& logger Nov 19 16:00:26 Tower logger: Starting Avahi mDNS/DNS-SD Daemon: /usr/sbin/avahi-daemon -D Nov 19 16:00:26 Tower avahi-daemon[1922]: Found user 'avahi' (UID 61) and group 'avahi' (GID 214). Nov 19 16:00:26 Tower avahi-daemon[1922]: Successfully dropped root privileges. Nov 19 16:00:26 Tower avahi-daemon[1922]: avahi-daemon 0.6.31 starting up. Nov 19 16:00:26 Tower avahi-daemon[1922]: WARNING: No NSS support for mDNS detected, consider installing nss-mdns! Nov 19 16:00:26 Tower avahi-daemon[1922]: Successfully called chroot(). Nov 19 16:00:26 Tower avahi-daemon[1922]: Successfully dropped remaining capabilities. Nov 19 16:00:26 Tower avahi-daemon[1922]: Loading service file /services/sftp-ssh.service. Nov 19 16:00:26 Tower avahi-daemon[1922]: Loading service file /services/smb.service. Nov 19 16:00:26 Tower avahi-daemon[1922]: Loading service file /services/ssh.service. Nov 19 16:00:26 Tower avahi-daemon[1922]: Joining mDNS multicast group on interface eth0.IPv4 with address 192.168.0.18. Nov 19 16:00:26 Tower avahi-daemon[1922]: New relevant interface eth0.IPv4 for mDNS. Nov 19 16:00:26 Tower avahi-daemon[1922]: Network interface enumeration completed. Nov 19 16:00:26 Tower avahi-daemon[1922]: Registering new address record for 192.168.0.18 on eth0.IPv4. Nov 19 16:00:26 Tower avahi-daemon[1922]: Registering HINFO record with values 'X86_64'/'LINUX'. Nov 19 16:00:26 Tower emhttp: shcmd (74): /etc/rc.d/rc.avahidnsconfd start |& logger Nov 19 16:00:26 Tower logger: Starting Avahi mDNS/DNS-SD DNS Server Configuration Daemon: /usr/sbin/avahi-dnsconfd -D Nov 19 16:00:26 Tower avahi-dnsconfd[1932]: Successfully connected to Avahi daemon. Nov 19 16:00:26 Tower emhttp: Starting Docker... Nov 19 16:00:26 Tower logger: Not starting Docker: no image file defined Nov 19 16:00:27 Tower avahi-daemon[1922]: Server startup complete. Host name is Tower.local. Local service cookie is 2744038094. Nov 19 16:00:28 Tower avahi-daemon[1922]: Service "Tower" (/services/ssh.service) successfully established. Nov 19 16:00:28 Tower avahi-daemon[1922]: Service "Tower" (/services/smb.service) successfully established. Nov 19 16:00:28 Tower avahi-daemon[1922]: Service "Tower" (/services/sftp-ssh.service) successfully established. I don't know what's causing this if this is the cause of my inability to write to the server. Any help or pointers gratefully received thanks! Paul syslog.txt
trurl Posted November 19, 2015 Posted November 19, 2015 sorry thought I'd done it..... No you only posted the syslog. For V6 we always want the complete diagnostics zip you get by going to Tools - Diagnostics
pm1961 Posted November 19, 2015 Author Posted November 19, 2015 Aha....... Thank you! tower-diagnostics-20151119-1622.zip
trurl Posted November 19, 2015 Posted November 19, 2015 None of your drives SMART is reported in the diagnostics. I suspect something to do with your controller. What is your hardware? The portion of your syslog when it tries to mount your cache drive seems especially odd Nov 19 16:00:24 Tower emhttp: shcmd (58): mkdir -p /mnt/cache Nov 19 16:00:24 Tower emhttp: shcmd (59): set -o pipefail ; mount -t ntfs -o noatime,nodiratime /dev/sdo1 /mnt/cache |& logger Nov 19 16:00:24 Tower kernel: ntfs: driver 2.1.32 [Flags: R/O MODULE]. Nov 19 16:00:24 Tower logger: mount: warning: /mnt/cache seems to be mounted read-only. Nov 19 16:00:24 Tower kernel: ntfs: volume version 3.1. Nov 19 16:00:24 Tower emhttp: shcmd (60): sync Nov 19 16:00:26 Tower emhttp: shcmd (61): mkdir /mnt/user0 Nov 19 16:00:26 Tower emhttp: shcmd (62): /usr/local/sbin/shfs /mnt/user0 -disks 8190 -o noatime,big_writes,allow_other |& logger Nov 19 16:00:26 Tower emhttp: shcmd (63): mkdir /mnt/user Nov 19 16:00:26 Tower emhttp: shcmd (64): /usr/local/sbin/shfs /mnt/user -disks 8191 2048000000 -o noatime,big_writes,allow_other -o remember=0 |& logger Nov 19 16:00:26 Tower emhttp: shcmd (65): cat - > /boot/config/plugins/dynamix/mover.cron <<< "# Generated mover schedule:#01240 3 * * * /usr/local/sbin/mover |& logger#012" Nov 19 16:00:26 Tower emhttp: shcmd (66): /usr/local/sbin/update_cron &> /dev/null Nov 19 16:00:26 Tower emhttp: shcmd (67): :>/etc/samba/smb-shares.conf Nov 19 16:00:26 Tower emhttp: shcmd (68): chmod 0777 '/mnt/user/System Volume Information' Nov 19 16:00:26 Tower shfs/user: shfs_chmod: chmod: /mnt/cache/System Volume Information 040777 (30) Read-only file system Nov 19 16:00:26 Tower emhttp: shcmd: shcmd (68): exit status: 1 "ntfs" and "System Volume Information" makes me think this is somehow a disk from Windows but I wouldn't think unRAID would actually try to use NTFS volume as a cache drive. What is the history of your cache disk? Has it ever been in a Windows system?
pm1961 Posted November 19, 2015 Author Posted November 19, 2015 Thanks for the reply. The controller card has never reported SMART info, even when the system was behaving perfectly well. It's a 3Ware 9650SE 24 port. Apparently it's a known problem with the Linux 3Ware 9000 series driver? The cache drive is a recent thing. I thought I'd try it as the behaviour I described pre-dates the cache drive inclusion. Yes, it has been in a windows system but it was formatted before use. I shall try again without the cache drive and capture the data. Regards, Paul
pm1961 Posted November 20, 2015 Author Posted November 20, 2015 Well, removing the cache drive seems to have done the trick! I can now write to the RAID again. Thank you! That does leave me without a cache disk though. Writing directly to the RAID only chugs along at around an average of 5MB/S. Any thoughts about the way forward with this are most welcome. Regards, Paul
trurl Posted November 20, 2015 Posted November 20, 2015 Well, removing the cache drive seems to have done the trick! I can now write to the RAID again. Thank you! That does leave me without a cache disk though. Writing directly to the RAID only chugs along at around an average of 5MB/S. Any thoughts about the way forward with this are most welcome. Regards, Paul Is that speed writing over the network? If so I suspect a network issue. Your diagnostics seems to indicate it is OK on the unRAID side so maybe check cables, switches, or try from a different computer.
trurl Posted November 20, 2015 Posted November 20, 2015 Normally you will get an initial burst due to I/O caching in memory. Once memory is used up it will have to wait for the disks so it will slow down. Shouldn't be as slow as you are getting though. You could try going to the command line or using mc to copy between drives on the server and so eliminate the network to see if disk I/O by itself is better.
pm1961 Posted November 20, 2015 Author Posted November 20, 2015 Thanks for that. I'll start up a new topic since the aim of the title seems to have been sorted. Paul
Recommended Posts
Archived
This topic is now archived and is closed to further replies.