ken-ji

I think I solved it. I absolutely needed to add (sec=sys) For a NFS share called Test rule: empty /etc/exports: "/mnt/user/Test" -async,no_subtree_check,fsid=110 exportfs: exportfs: No host name given with /mnt/user/Test (ro,async,wdelay,hide,nocrossmnt,secure,root_squash,no_all_squash,no_subtree_check,secure_locks,acl,fsid=110,anonuid=65534,anongid=65534,sec=sys,ro,root_squash,no_all_squash), suggest *(ro,async,wdelay,hide,nocrossmnt,secure,root_squash,no_all_squash,no_subtree_check,secure_locks,acl,fsid=110,anonuid=65534,anongid=65534,sec=sys,ro,root_squash,no_all_squash) to avoid warning /mnt/user/Test <world>(ro,async,wdelay,root_squash,no_subtree_check,fsid=110,sec=sys,ro,root_squash,no_all_squash) With a rule like: rule: 192.168.2.0/24(rw) /etc/exports: "/mnt/user/Test" -async,no_subtree_check,fsid=110 192.168.2.0/24(rw) exportfs: /mnt/user/Test 192.168.2.0/24(ro,async,wdelay,root_squash,no_subtree_check,fsid=110,sec=sys,ro,root_squash,no_all_squash) /mnt/disk1 <world>(ro,async,wdelay,insecure,root_squash,all_squash,no_subtree_check,fsid=11,anonuid=99,anongid=100,sec=sys,ro,root_squash,all_squash) the rw does not take effect unless I add sec=sys Rule: 192.168.2.4(sec=sys,rw) /etc/exports: "/mnt/user/Test" -async,no_subtree_check,fsid=110 192.168.2.0/24(sec=sys,rw) exportfs: /mnt/user/Test 192.168.2.0/24(rw,async,wdelay,root_squash,no_subtree_check,fsid=110,sec=sys,rw,root_squash,no_all_squash) /mnt/disk1 <world>(ro,async,wdelay,insecure,root_squash,all_squash,no_subtree_check,fsid=11,anonuid=99,anongid=100,sec=sys,ro,root_squash,all_squash) Hope this helps

Tom, where can we look into the current and future kernel config? That way we know if we will need to request or build our own modules now and in the future? I'm thinking of sticking an arduino as an LCD screen + USB HID to my unRAID server.

Will look into it for future versions Why is this needed? What is wrong with automatic detection? If you run lsscsi what do you get? Just hedging your bets with autodetection. but as for the WD Red 4Gs lssci: smartctl Note the emphasis and:

Maybe we can have arbitrary groups of disks? I have an external SAS enclosure and a few disks inside the case. A 2 column layout is wrong since the number of disks are not the same. Just a guess. Probably allow the user to mark the disks as SATA/ATA/etc. Odd that WD Reds of 4GB and higher are not in the smartctl db as of v6rc3

What does ls -las /dev/disk/by-id show? The disk list is built out of those devices that start with sata- (and don't contain -part). If your devices began ata- for example, they wouldn't be found.. Why are ata devices being skipped? Mine are all SATA devices (in a SAS Expander Tower)

Hi guys, Setting up my first unRaid box to replace my Slackware NAS (after I destroyed my data with an (unknowingly) aborted MD resize operation ) I'm using 6b14 with Pro license. However, I can't make sense of the NFS export option for user shares... Both Public and Secure give me: root@MediaStore:~# exportfs -v /mnt/user/Media <world>(rw,async,wdelay,insecure,root_squash,all_squash,no_subtree_check,fsid=100,anonuid=99,anongid=100,sec=sys,rw,root_squash,all_squash) root@MediaStore:~# exportfs -v /mnt/user/Media <world>(ro,async,wdelay,insecure,root_squash,all_squash,no_subtree_check,fsid=100,anonuid=99,anongid=100,sec=sys,ro,root_squash,all_squash) While private with 192.168.2.2(rw) gives me: root@MediaStore:~# exportfs -v /mnt/user/Media 192.168.2.2(ro,async,wdelay,insecure,root_squash,all_squash,no_subtree_check,fsid=100,anonuid=99,anongid=100,sec=sys,ro,root_squash,all_squash) So does this mean to have writable NFS shares, they can only be public? while secure is read-only and private is read-only to specific machines/networks?