Get a list of all disks and show the serial numbers
|
1 2 3 4 5 6 7 8 9 10 |
#!/bin/bash if [ ! -x /usr/sbin/hdparm ] ; then echo "Missing hdparm" exit 1 fi for disk in /dev/sd?; do echo "$disk: `hdparm -I $disk | grep "Serial Number:" | awk '{ print $3 }'`" done |
Another wrapper (like mdstat), that’ll look through my Western Digital disks and fix any disk not having the head parking timeout set to a configured value.
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 |
#!/bin/bash FIRMWARE_TIMEOUT_SH=138 if [ ! -x /usr/sbin/idle3ctl ] ; then echo "idle3-tools is missing." exit 1 fi for disk in /dev/sd?; do # idle time should be set to: 138 (300s) # 138-128=10 # 10x30=300 seconds FIRMWARE_TIMEOUT_IS="$( /usr/sbin/idle3ctl -g $disk | awk '{ print $5 }' )" if [ "$FIRMWARE_TIMEOUT_IS" -ne "$FIRMWARE_TIMEOUT_SH" ]; then /usr/sbin/idle3ctl -s$FIRMWARE_TIMEOUT_SH $disk fi done |
Well, I needed a way to watch the mdstat progress (because a disk just failed …).
|
1 2 3 |
#!/bin/bash watch -n1 cat /proc/mdstat |
Well, I have a few movies and series that ain’t represented in TMDB/TVDB. So here’s a little script, that will parse over any video files, check if a thumb file is already present, and if not generate one using ffmpegthumbnailer…
|
1 2 3 4 5 6 7 8 |
#!/bin/bash find /srv/smb/tv/ -name "*.wmv" -o -name "*.avi" -o -name "*.mp4" \ -o -name "*.mkv" | while read file; do if [ ! -f "${file%.*}-thumb.jpg" ] ; then fmpegthumbnailer -i $file -o "${file%.*}-thumb.jpg" -s 0 &>/dev/null fi done |
Well, I recently had to flatten my archive NAS (well only the OS part … *wheeeh*). Since I didn’t have the chance to backup the old settings I had to do everything from scratch … And this time I decided, I wasn’t doing a script but rather the proper way.
I spent a while reading through the Internetz about the various settings until I stumbled upon a Frauenhofer Wiki entry. From there I ended up writing those udev-rules and the sysctl configs…
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 |
# Settings from http://www.fhgfs.com/wiki/StorageServerTuning # Set an appropriate IO scheduler for file servers. KERNEL=="sd[a-z]", ATTR{queue/scheduler}="deadline" KERNEL=="sd[a-i][a-z]", ATTR{queue/scheduler}="deadline" # Give the IO scheduler more flexibility by increasing the number of # schedulable requests. KERNEL=="sd[a-z]", ATTR{queue/nr_requests}="4096" KERNEL=="sd[a-i][a-z]", ATTR{queue/nr_requests}="4096" # To improve throughput for sequential reads, increase the maximum amount of # read-ahead data. The actual amount of read-ahead is adaptive, so using a # high value here won't harm performance for small random access. KERNEL=="sd[a-z]", ATTR{queue/read_ahead_kb}="73728" KERNEL=="sd[a-i][a-z]", ATTR{queue/read_ahead_kb}="73728" KERNEL=="sd[a-z]", RUN+="/sbin/blockdev --setra 73728 /dev/%n" KERNEL=="sd[a-i][a-z]", RUN+="/sbin/blockdev --setra 73728 /dev/%n" SUBSYSTEM=="block", KERNEL=="md[0-9]*", RUN+="/sbin/blockdev --setra 663552 /dev/%n" SUBSYSTEM=="block", KERNEL=="md[0-9]*", ATTR{md/stripe_cache_size}="9216" # Optimal performance for hardware RAID systems often depends on large IOs # being sent to the device in a single large operation. Please refer to your # hardware storage vendor for the corresponding optimal size of # /sys/block/sdX/max_sectors_kb. # It is typically good if this size can be increased to at least match your # RAID stripe set size (i.e. chunk_size x number_of_disks): KERNEL=="sd[a-z]", ATTR{queue/max_sectors_kb}="512" KERNEL=="sd[a-i][a-z]", ATTR{queue/max_sectors_kb}="512" KERNEL=="sd[a-z]", ATTR{device/queue_depth}="1" KERNEL=="sd[a-i][a-z]", ATTR{device/queue_depth}="1" |
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 |
# Settings taken from http://www.fhgfs.com/wiki/StorageServerTuning # To avoid long IO stalls (latencies) for write cache flushing in a production # environment with very different workloads, you will typically want to limit # the kernel dirty (write) cache size. vm.dirty_background_ratio = 5 vm.dirty_ratio = 10 # Assigning slightly higher priority to inode caching helps to avoid disk seeks # for inode loading vm.vfs_cache_pressure = 50 # Buffering of file system data requires frequent memory allocation. Raising the # amount of reserved kernel memory will enable faster and more reliable memory # allocation in critical situations. Raise the corresponding value to 64MB if # you have less than 8GB of memory, otherwise raise it to at least 256MB vm.min_free_kbytes = 262144 |
For now, I’m rather pleased with the results …
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 |
root:(charon.ka.heimdaheim.de) PWD:/ Wed Jul 09, 15:02:08 [0] > mdadm --detail /dev/md127 /dev/md127: Version : 1.2 Creation Time : Sat Jan 26 18:35:19 2013 Raid Level : raid5 Array Size : 15626121216 (14902.23 GiB 16001.15 GB) Used Dev Size : 1953265152 (1862.78 GiB 2000.14 GB) Raid Devices : 9 Total Devices : 10 Persistence : Superblock is persistent Update Time : Wed Jul 9 15:03:28 2014 State : clean Active Devices : 9 Working Devices : 10 Failed Devices : 0 Spare Devices : 1 Layout : left-symmetric Chunk Size : 512K Name : charon:aggr1 (local to host charon) UUID : 6d11820f:04847070:2725c434:9ee39718 Events : 11186 Number Major Minor RaidDevice State 0 8 129 0 active sync /dev/sdi1 1 8 33 1 active sync /dev/sdc1 2 8 49 2 active sync /dev/sdd1 4 8 65 3 active sync /dev/sde1 5 8 17 4 active sync /dev/sdb1 10 8 97 5 active sync /dev/sdg1 9 8 81 6 active sync /dev/sdf1 8 8 161 7 active sync /dev/sdk1 7 8 145 8 active sync /dev/sdj1 6 8 113 - spare /dev/sdh1 |
And here’s the dd output:
|
1 2 3 4 5 6 |
root:(charon.ka.heimdaheim.de) PWD:/ Wed Jul 09, 14:57:32 [0] > dd if=/dev/zero of=/srv/smb/tmp bs=1G count=100 \ oflag=direct 100+0 records in 100+0 records out 107374182400 bytes (107 GB) copied, 257.341 s, 417 MB/s |