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 |
Since I’m living and working in Germany, most of my hardware is using timezone configuration for CEST (or Europe/Berlin).
This here is the simple configuration for our Nexus 5000’s:
|
1 2 3 4 5 |
conf t clock timezone CET 1 0 clock summer-time CEST 5 Sun Mar 02:00 5 Sun Oct 03:00 60 exit copy running-config startup-config |
For the past two months we’ve been trying to migrate a bunch (90 or so) VMs from XenServer to ESXi … However for some reason on some of them, the Converter Service would crash.
Up till Monday, I had no idea why. I decided to look into the error once again, and this time decided just to Google the failing module… And guess what ? Out came this Citrix forum post regarding the failing module. So, after knowing that rsintcor32.dll belongs to the Citrix System Monitoring Agent service (well, I could have guessed that from the DLLs path :P) I decided to simply stop the service.
And now, we can migrate the remaining VMs to ESXi and get rid of XenServer!
One of my co-workers recently approached me, that he needed a simple shell script which would generate a simple report about a Vserver’s current connections. After ironing out a few things with him (he had the intention of it being on a CIFS share on our file-server – which I changed to a simple HTML page) I went to work.
Out came two scripts. One is the collection instance, and the other is the processing instance. First the collection script runs, finds the current HA master node and then collects the Vserver’s current connections. After that script has dumped the information (date, time, current connections) into a file, the processing script will go and create a simple HTML page that’ll show exactly those informations.
You’ll also need to have configured the public key authentification on both HA nodes, since entering a password in combination with scripting is a bit lame.
|
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 39 40 |
#!/bin/bash IDENTITY_FILE="/root/.ssh/netscaler" NS_USER="nsroot" NS_HOST="$1" NS_VSERVER="$2" if [ -z "$2" ] ; then echo "Usage: ns-gather-connections.sh <IP/HOSTNAME> <VSERVER_NAME>" echo "" echo " o IP/HOSTNAME - IP or Hostname of the NetScaler" echo " o VSERVER_NAME - Name of the NetScaler vServer" echo "" exit 1 fi ns_ssh() { COMMANDS="$@" ssh -i $IDENTITY_FILE $NS_USER@$NS_HOST $COMMANDS } # Get HA information from the NetScaler and check if the current IP is the Master NS_NODE_STATUS="$( ns_ssh sh ha node | egrep -A2 "IP:.*$NS_HOST" | grep Master | \ awk '{ print $3 }' )" if [ "$NS_NODE_STATUS" != "Primary" ] ; then # Search for the Primary node NS_MASTER_NODE="$( ns_ssh sh ha node | egrep -B2 "Master State: Primary" | \ grep "IP: " | awk '{ print $2 }' )" NS_HOST=$NS_MASTER_NODE elif [ "$NS_NODE_STATUS" == "Primary" ] ; then NS_MASTER_NODE="$NS_HOST" NS_HOST=$NS_MASTER_NODE fi # Check the connection table NS_CONNTABLE="$( ns_ssh "sh ns connectiontable \"VSVRNAME = ${NS_VSERVER}\" -detail LINK" | grep 'ESTABLISHED' | wc -l )" echo "`date '+%F_%R'`;$NS_CONNTABLE" >> /srv/www/${NS_VSERVER}_conntable.log |
|
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 |
#!/bin/bash NS_VSERVER="$1" HTML_FILE="/srv/www/htdocs/${NS_VSERVER}.html" cat > $HTML_FILE << EOF <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> <html> <head> <title>$NS_VSERVER - Anzahl Clients nach Datum</title> <style> .ns { margin:0px;padding:0px; width:100%; border:1px solid #000000; -moz-border-radius-bottomleft:0px; -webkit-border-bottom-left-radius:0px; border-bottom-left-radius:0px; -moz-border-radius-bottomright:0px; -webkit-border-bottom-right-radius:0px; border-bottom-right-radius:0px; -moz-border-radius-topright:0px; -webkit-border-top-right-radius:0px; border-top-right-radius:0px; -moz-border-radius-topleft:0px; -webkit-border-top-left-radius:0px; border-top-left-radius:0px; }.ns table{ border-collapse: collapse; border-spacing: 0; width:100%; height:100%; margin:0px;padding:0px; }.ns th{ font-size:14px; font-family:Verdana; font-weight:bold; border:1px solid #000000; border-width:0px 1px 1px 0px; text-align:left; padding:7px; }.ns tr:last-child td:last-child { -moz-border-radius-bottomright:0px; -webkit-border-bottom-right-radius:0px; border-bottom-right-radius:0px; } .ns table tr:first-child td:first-child { -moz-border-radius-topleft:0px; -webkit-border-top-left-radius:0px; border-top-left-radius:0px; } .ns table tr:first-child td:last-child { -moz-border-radius-topright:0px; -webkit-border-top-right-radius:0px; border-top-right-radius:0px; }.ns tr:last-child td:first-child{ -moz-border-radius-bottomleft:0px; -webkit-border-bottom-left-radius:0px; border-bottom-left-radius:0px; }.ns tr:hover td{ } .ns tr:nth-child(odd){ background-color:#aad4ff; } .ns tr:nth-child(even) { background-color:#ffffff; }.ns td{ vertical-align:middle; border:1px solid #000000; border-width:0px 1px 1px 0px; text-align:left; padding:7px; font-size:14px; font-family:Verdana; font-weight:normal; color:#000000; }.ns tr:last-child td{ border-width:0px 1px 0px 0px; }.ns tr td:last-child{ border-width:0px 0px 1px 0px; }.ns tr:last-child td:last-child{ border-width:0px 0px 0px 0px; } .ns tr:first-child td{ background:-o-linear-gradient(bottom, #005fbf 5%, #003f7f 100%); background:-webkit-gradient( linear, left top, left bottom, color-stop(0.05, #005fbf), color-stop(1, #003f7f) ); background:-moz-linear-gradient( center top, #005fbf 5%, #003f7f 100% ); filter:progid:DXImageTransform.Microsoft.gradient(startColorstr="#005fbf", endColorstr="#003f7f"); background: -o-linear-gradient(top,#005fbf,003f7f); background-color:#005fbf; border:0px solid #000000; text-align:center; border-width:0px 0px 1px 1px; font-size:14px; font-family:Verdana; font-weight:bold; color:#ffffff; } .ns tr:first-child:hover td{ background:-o-linear-gradient(bottom, #005fbf 5%, #003f7f 100%); background:-webkit-gradient( linear, left top, left bottom, color-stop(0.05, #005fbf), color-stop(1, #003f7f) ); background:-moz-linear-gradient( center top, #005fbf 5%, #003f7f 100% ); filter:progid:DXImageTransform.Microsoft.gradient(startColorstr="#005fbf", endColorstr="#003f7f"); background: -o-linear-gradient(top,#005fbf,003f7f); background-color:#005fbf; } .ns tr:first-child td:first-child{ border-width:0px 0px 1px 0px; } .ns tr:first-child td:last-child{ border-width:0px 0px 1px 1px; } </style> </head> <body> <h3>NetScaler Vserver $NS_VSERVER - sortiert nach Datum</h1> <div class="ns"> <table border="1"> <tr> <th>Datum</th> <th>Zeit</th> <th>Anzahl Clients</th> </tr> EOF for line in $( tac /srv/www/${NS_VSERVER}_conntable.log ); do DATE="$( echo $line | cut -d\; -f1 | cut -d_ -f1 )" TIME="$( echo $line | cut -d\; -f1 | cut -d_ -f2 )" CONNECTIONS="$( echo $line | cut -d\; -f2 )" cat >> $HTML_FILE << EOF <tr> <td>$DATE</td> <td>$TIME</td> <td>$CONNECTIONS</td> </tr> EOF done cat >> $HTML_FILE << EOF </div> </body> </html> EOF |
Well, I’ve been fiddling around with SLES and openSUSE VMware templates. I know it’s a stupid idea when you have a PXE server from which you could install this in a matter of minutes (seriously the SLES PXE installation takes about 5 minutes).
However, when dealing with DMZ’s (yeah, they exist!) you usually don’t have any PXE servers there. So I decided to go with simple VMware templates (like we do with Windows already), but had to iron out a few kinks.
So of I went and wrote a short (70 line ..) init script, which will do exactly that.
|
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 |
#!/bin/sh # # /etc/init.d/template-deploy # ### BEGIN INIT INFO # Provides: template-deploy # Required-Start: $sshd # Should-Start: # Required-Stop: $sshd # Should-Stop: # Default-Start: 2 3 5 # Default-Stop: # Short-Description: Makes the final adjustments when deployed a template # Description: VMware has a few kinks when deploying a linux template. # This init script will handle those kinks and # incorporate the necessary changes to the final system # after the VMware customizations are done. ### END INIT INFO # Source LSB init functions . /etc/rc.status rc_reset LOGFILES="/var/log/messages* /var/log/mail* /var/log/zypper.log /var/log/localmessages /var/log/warn /var/log/NetworkManager /var/log/zypp/history /var/log/clamd.log /var/log/cksquid.log /var/log/ftp-proxy.log* /var/log/apache2/*_log /var/log/squid/*.log /var/log/pbl.log" case "$1" in start) # Read /etc/HOSTNAME and check it against /etc/template if [ -f /etc/HOSTNAME -a -f /etc/template ] ; then SYS_HOSTNAME="`cat /etc/HOSTNAME`" TPL_HOSTNAME="`cat /etc/template`" # Only if the hostname isn't like the template name # the script will actually run. if [ "$SYS_HOSTNAME" != "$TPL_HOSTNAME" ] ; then sed -i "s/$TPL_HOSTNAME/$SYS_HOSTNAME/g" \ /etc/postfix/main.cf /etc/hosts \ /etc/HOSTNAME bash /var/adm/autoinstall/scripts/configure_users.sh # Stop services for logfile deletion. systemctl stop postfix systemctl stop syslog # Remove log files for file in $LOGFILES; do rm -f $file done rm -f /var/log/YaST2/* rm -rf /var/log/vmware-imc # Remove the init script chkconfig -d template-deploy rm -f /etc/template rm -f /etc/init.d/template-deploy fi fi rc_status -v ;; *) ;; esac rc_exit |
You’ll also need to create a file (/etc/template), which’ll hold the template’s hostname and will be used for the comparison if VMware’s post-processing is already finished.
Well, I went to work today … yeah, I know it’s Sunday right ? I ended up updating two MDS9148 switches and I didn’t want to figure out everything all over again. So I put the system image and kickstart onto one of our FTP servers and ran a short bash loop on it:
|
1 2 3 4 |
root:(ftp.daheim.heimdaheim.de) PWD:~ Sun Oct 20, 08:57:24 [0] > for file in `ls /srv/ftp/firmware/mds9148/*6.2.3*`; do echo "copy ftp://10.0.0.55/`echo $file | sed 's,/srv/ftp/,,'` bootflash:/`echo $file | sed 's,/srv/ftp/firmware/mds9148/,,'`" done |
Now that’ll generate me two lines, which in turn I can use on the MDS’n:
|
1 2 |
copy ftp://10.0.0.55/firmware/mds9148/m9100-s3ek9-kickstart-mz.6.2.3.bin bootflash:/m9100-s3ek9-kickstart-mz.6.2.3.bin copy ftp://10.0.0.55/firmware/mds9148/m9100-s3ek9-mz.6.2.3.bin bootflash:/m9100-s3ek9-mz.6.2.3.bin |