Mark's blog - Comments

Rene: Linux, Solaris and FreeBSD iostat monitoring with Cacti

nospam@example.com (Rene) — Thu, 13 Oct 2011 10:06:04 +0100

Hello, thank you very much for this awesome tool I experienced some weird error messages while performing an "verbose" snmp-query @cacti. the error message was: "+ Invalid field ioDescr:ioName:ioIndex" "+ Must contain input fields only" to resolve the issue i've modified the iostat.xml file (as christianha mentioned above). changed from: ioDescr:ioName:ioIndex to: ioDescr:ioIndex just wanted to "archive" the error-message, since i didn't find anything via google. keep up the great work best regards, René

Sean: Linux, Solaris and FreeBSD iostat monitoring with Cacti

nospam@example.com (Sean) — Wed, 12 Oct 2011 16:55:45 +0100

I'm experiencing some strange behavior on certain servers that in all aspects appear the same as servers that have no issue. I don't believe it's the script, it seems like its SNMP related somehow. For instance, iostat.cache on a working machine: Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await svctm %util sda 0.51 95.76 3.71 24.35 122.43 480.47 42.96 0.06 2.28 0.43 1.21 sda1 0.00 0.00 0.00 0.00 0.00 0.00 12.45 0.00 6.39 6.21 0.00 sda2 0.51 95.76 3.71 24.35 122.43 480.47 42.96 0.06 2.28 0.43 1.21 sda3 0.00 0.00 0.00 0.00 0.00 0.00 53.71 0.00 4.60 4.07 0.00 Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await svctm %util sda 0.20 72.30 1.03 29.07 38.80 405.47 29.52 0.03 1.06 0.30 0.89 sda1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 sda2 0.20 72.30 1.03 29.07 38.80 405.47 29.52 0.03 1.06 0.30 0.89 sda3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 iostat.cache on a broken machine: Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await svctm %util sda 0.10 8.49 0.46 1.42 11.11 39.61 54.10 0.02 8.52 0.33 0.06 sda1 0.00 0.00 0.00 0.00 0.00 0.00 23.25 0.00 2.74 1.80 0.00 sda2 0.09 8.49 0.46 1.42 11.10 39.61 54.10 0.02 8.52 0.33 0.06 sda3 0.00 0.00 0.00 0.00 0.00 0.00 48.97 0.00 6.12 5.73 0.00 Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await svctm %util sda 0.00 2.07 0.00 1.83 0.00 15.60 17.02 0.00 0.44 0.05 0.01 sda1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 sda2 0.00 2.07 0.00 1.83 0.00 15.60 17.02 0.00 0.44 0.05 0.01 sda3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 The problem is that snmpwalk on the working machine ends promptly with this: [root@adm01b log]# snmpwalk -v2c -cpublic web05b .1.3.6.1.3.1 | tail SNMPv2-SMI::experimental.1.11.3 = STRING: "0.80" SNMPv2-SMI::experimental.1.11.4 = STRING: "0.00" SNMPv2-SMI::experimental.1.12.1 = STRING: "0.35" SNMPv2-SMI::experimental.1.12.2 = STRING: "0.00" SNMPv2-SMI::experimental.1.12.3 = STRING: "0.35" SNMPv2-SMI::experimental.1.12.4 = STRING: "0.00" SNMPv2-SMI::experimental.1.13.1 = STRING: "1.10" SNMPv2-SMI::experimental.1.13.2 = STRING: "0.00" SNMPv2-SMI::experimental.1.13.3 = STRING: "1.10" SNMPv2-SMI::experimental.1.13.4 = STRING: "0.00" Non-working machine: [root@adm01b log]# snmpwalk -v2c -cpublic web65b .1.3.6.1.3.1 SNMPv2-SMI::experimental.1.11.3 = STRING: "0.40" SNMPv2-SMI::experimental.1.11.4 = STRING: "0.00" SNMPv2-SMI::experimental.1.12.1 = STRING: "0.06" SNMPv2-SMI::experimental.1.12.2 = STRING: "0.00" SNMPv2-SMI::experimental.1.12.3 = STRING: "0.06" SNMPv2-SMI::experimental.1.12.4 = STRING: "0.00" SNMPv2-SMI::experimental.1.13.1 = STRING: "0.01" SNMPv2-SMI::experimental.1.13.2 = STRING: "0.00" SNMPv2-SMI::experimental.1.13.3 = STRING: "0.01" SNMPv2-SMI::experimental.1.13.4 = STRING: "0.00" Timeout: No Response from web65b Running it in debug is where I see the difference in responses on the last MiB: working: SNMPv2-SMI::experimental.1.13.4 = STRING: "0.00" Sending 42 bytes to UDP: [192.168.82.5]:161 0000: 30 28 02 01 01 04 06 70 75 62 6C 69 63 A1 1B 02 0(.....public... 0016: 04 10 47 C3 C5 02 01 00 02 01 00 30 0D 30 0B 06 ..G........0.0.. 0032: 07 2B 06 01 03 01 0D 04 05 00 .+........ Received 42 bytes from UDP: [192.168.82.5]:161 0000: 30 28 02 01 01 04 06 70 75 62 6C 69 63 A2 1B 02 0(.....public... 0016: 04 10 47 C3 C5 02 01 00 02 01 00 30 0D 30 0B 06 ..G........0.0.. 0032: 06 2B 06 01 03 02 01 41 01 74 .+.....A.t NON-working: SNMPv2-SMI::experimental.1.13.4 = STRING: "0.00" Sending 42 bytes to UDP: [192.168.82.65]:161 0000: 30 28 02 01 01 04 06 70 75 62 6C 69 63 A1 1B 02 0(.....public... 0016: 04 18 0F 5C 3A 02 01 00 02 01 00 30 0D 30 0B 06 ...\:......0.0.. 0032: 07 2B 06 01 03 01 0D 04 05 00 .+........ Resending 42 bytes to UDP: [192.168.82.65]:161 0000: 30 28 02 01 01 04 06 70 75 62 6C 69 63 A1 1B 02 0(.....public... 0016: 04 18 0F 5C 3A 02 01 00 02 01 00 30 0D 30 0B 06 ...\:......0.0.. 0032: 07 2B 06 01 03 01 0D 04 05 00 .+........ (resends 4x then times out) Note the lack of response containing ".A.t" from the broken machine. Do you have any idea what I should do next for troubleshooting?

Alexander: LigHTTPd and Apache - Symfony benchmarks

nospam@example.com (Alexander) — Wed, 12 Oct 2011 10:12:37 +0100

Good article about ApacheBench http://plutov.by/post/apache_bench_easy_tool

Goncalo Gomes: Citrix XenServer 5.6 Review

nospam@example.com (Goncalo Gomes) — Thu, 06 Oct 2011 23:21:28 +0100

That's actually false. You may install additional software, but if you run into any problems later and decide to engage XenServer support, the support engineer may tell you he is going to support your configuration in 'best effort'. Citrix definition of best effort is: We will support you as long as we believe the problem you are experiencing is not a side effect of one of the unsupported extras you've thrown into your configuration. While I can only speak for Citrix, I am sure this is possibly standard across all enterprise software, including microsoft, oracle, vmware, sap, etc. Thanks for the article. Nicely covered.

Darren Murphy: Linux, Solaris and FreeBSD iostat monitoring with Cacti

nospam@example.com (Darren Murphy) — Sun, 28 Aug 2011 15:09:35 +0100

Hi Mark, You may have already noticed, but in case you didn't.... I've forked your code on GitHub and made a few mods, mainly to suit my own purposes. regards, Darren

Michael Laursen: Building a redundant iSCSI and NFS cluster with Debian - Part 4

nospam@example.com (Michael Laursen) — Tue, 23 Aug 2011 22:07:30 +0100

I seem to have run into yet another obstacle I have two nodes, SAN02 and SAN03. Everything works fine untill I kill SAN02. Initiator looses connection to the underlying iSCSI disk when I so. cat /proc/net/iet/volumes shows that it migrated to SAN03 correctly. But client cannot connect to the drive. If I run /etc/init.d/iscsitarget restart on SAN03 while SAN02 is still down, the client reconnects to the drive. I am puzzled by this behaviour. I cannot manually go do this each time something happens to one of the nodes. Any ideas?

Michael Laursen: Building a redundant iSCSI and NFS cluster with Debian - Part 4

nospam@example.com (Michael Laursen) — Mon, 22 Aug 2011 23:44:37 +0100

I am having trouble as to the part where heartbeat should "bring up" eth2 and eth3 with the ip-addresses specified in haresources. I have not configured these interfaces with any ip's on any of the servers through /etc/network/interfaces. Is this correct? If not, how would I configure them? Thanks in advance. Great article!

TS: Xenserver snapshot and template based backup script

nospam@example.com (TS) — Mon, 22 Aug 2011 23:19:22 +0100

I'm trying to restore from the template, but not having luck. When the restored VM boots I get "Boot device: Hard Drive - failure: disk not bootable" What could I be doing wrong?

Joey Trungale: Linux, Solaris and FreeBSD iostat monitoring with Cacti

nospam@example.com (Joey Trungale) — Tue, 09 Aug 2011 07:28:45 +0100

If you want to save yourself some clicks, make use of the cacti cli when importing templates: for i in /path/to/Cacti-iostat-templates/templates/linux/cacti_*.xml do php ./import_template.php -filename=$i done

Alex: Apache mod_proxy balancing with PHP sticky sessions

nospam@example.com (Alex) — Sun, 26 Jun 2011 21:29:02 +0100

I got mine working like this... ServerAdmin webmaster@localhost ProxyRequests Off Order deny,allow Allow from all ProxyPass /error1 http://192.168.1.41/error/ ProxyPass /error2 http://192.168.1.42/error/ ProxyPass /balancer-manager ! SetHandler balancer-manager Order deny,allow Allow from 192.168.1.0/24 Allow from 10.1.0.0/24 Header add Set-Cookie "ROUTEID=.%{BALANCER_WORKER_ROUTE}e; path=/" env=BALANCER_ROUTE_CHANGED BalancerMember http://192.168.1.41:80 route=1 BalancerMember http://192.168.1.42:80 route=2 #ProxySet lbmethod=byrequests ProxySet stickysession=ROUTEID ProxyPass / balancer://cluster/ nofailover=On #stickysession=BALANCEID ProxyPassReverse / balancer://cluster/ Options FollowSymLinks AllowOverride None ErrorLog /var/log/apache2/error.log # Possible values include: debug, info, notice, warn, error, crit, # alert, emerg. LogLevel warn CustomLog /var/log/apache2/access.log combined CookieTracking on CookieStyle Cookie CookieName USERINFO # LogSQLTransferLogFormat huSUsbTvRAc LogSQLWhichCookie USERINFO LogSQLTransferLogTable access_log LogSQLCookieLogTable cookies LogSQLCreateTables on Does anynyone know how to redirect to working route if one route fails?

James Liu: Linux, Solaris and FreeBSD iostat monitoring with Cacti

nospam@example.com (James Liu) — Sun, 26 Jun 2011 09:04:13 +0100

Hi Mark, Thanks for your fantastic work. I have an issue about no snmp data returns. My /tmp/iostat.cache file is : Linux 2.6.18-194.el5 (n3.db.l99.com) 06/26/11 x86_64 (24 CPU) Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await svctm %util cciss/c0d0 0.08 7.67 0.36 2.41 18.93 51.64 50.87 0.00 0.64 0.25 0.07 dm-0 0.00 0.00 0.44 10.08 18.93 51.64 13.42 0.01 1.09 0.06 0.07 dm-1 0.00 0.00 0.00 0.00 0.00 0.00 8.00 0.00 4.13 0.33 0.00 Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await svctm %util cciss/c0d0 0.00 0.90 0.00 1.10 0.00 8.00 14.55 0.00 0.36 0.03 0.00 dm-0 0.00 0.00 0.00 2.00 0.00 8.00 8.00 0.00 0.25 0.02 0.00 dm-1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 And disabled SELinux , returns when I use snmpwalk command (.1.3.6.1.3.1.2) is below: SNMPv2-SMI::experimental.1.2.1 = STRING: "cciss/c0d0" SNMPv2-SMI::experimental.1.2.2 = STRING: "dm-0" SNMPv2-SMI::experimental.1.2.3 = STRING: "dm-1" When I add query to my device in Cacti 0.8.7g, I get these debug infos: + Running data query [20]. + Found type = '3' [snmp query]. + Found data query XML file at '/usr/local/cacti/resource/snmp_queries/iostat.xml' + XML file parsed ok. + Executing SNMP walk for list of indexes @ '.1.3.6.1.3.1' + No SNMP data returned + Found data query XML file at '/usr/local/cacti/resource/snmp_queries/iostat.xml' + Found data query XML file at '/usr/local/cacti/resource/snmp_queries/iostat.xml' + Found data query XML file at '/usr/local/cacti/resource/snmp_queries/iostat.xml' I'm stuck , plz help me get out of this. Thanks a lot.

L0cutus: ZFS Replication

nospam@example.com (L0cutus) — Thu, 05 May 2011 08:24:35 +0100

so is this script to backup my /home ok ?: http://pastebin.com/cuP7MbJX thanks !

Indiana Jones: Linux, Solaris and FreeBSD iostat monitoring with Cacti

nospam@example.com (Indiana Jones) — Thu, 28 Apr 2011 21:36:40 +0100

Did you check to see if the temp file is created? Did you walk the OID directly? Did you add the config in the snmp config?

Mark Round: Xenserver snapshot and template based backup script

nospam@example.com (Mark Round) — Thu, 28 Apr 2011 13:27:33 +0100

Yes please - that would be very useful!

Luis Davim: Xenserver snapshot and template based backup script

nospam@example.com (Luis Davim) — Thu, 28 Apr 2011 13:12:37 +0100

Hello I've made some modifications to this script to allow backing up the VMs to xva files with an independent schedule of the template backups, I use this to store xva files of my VMs in tapes monthly. If you are interested in my version of the script I can send it to you...