Posts

Showing posts from 2014

Extending a LVM logical volume with SaltStack

Image
How do you, at once, extend a LVM logical volume on a fleet of identical linux ( Centos ) servers using SaltStack ? Here's how and, thanks to Salt, it only took 5m.

Indexing Apache access logs with ELK (Elasticsearch+Logstash+Kibana)

Image
Who said that grepping Apache logs has to be boring? Sample of dashboard that can be created with ELK. Pretty impressive, huh? The truth is that, as Enteprise applications move to the browser too, Apache access logs are a gold mine, it does not matter what your role is: developer, support or sysadmin. If you are not mining them you are most likely missing out a ton of information and, probably, making the wrong decisions. ELK (Elasticsearch, Logstash, Kibana) is a terrific, Open Source stack for visually analyzing Apache (or nginx) logs (but also any other timestamped data).

Extract TABLE data from a large postgres SQL dump (with postgis)

What do you do when postgres refuses to import a dump because it contains invalid byte sequences? Solution: feed the sql script to iconv then import it as usual. That's easier said than done especially if your database contains postgis data which must be restored through a custom postgres dump (instructions here ). I recently experienced this issue on a relatively small table in a large-ish database. Since hand editing the SQL dump is cumbersome and hard (it is over 500MB in size) the only and most elegant alternative was to do it with a script. The following is an awk script which will extract the COPY instructions relative to a table from a postgres SQL dump: File: copy_estract.awk ---------------------- BEGIN {start=0} /^COPY "/ { if(index($0,TBL)!=0) { start=1; } } // {if(start==1) print $0;} /\\\./ {start=0;} Usage: awk -f copy_extract.awk -v TBL=TABLENAME pgdump/database_dump.sql One liner: awk -f copy_extract.awk -v TBL=TEST pgdump/d...

Ehcache: deploy multiple versions of a Grails app (fix javax.management.InstanceAlreadyExistsException)

When a Grails application makes use of the Ehcache cache plugin in its default configuration it can be impossibile to perform deploys of multiple versions of the app, even though the container might support it. The same plugin (in its default configuration) also breaks deploying multiple different Grails apps on the same container. The problem is in the way the plugin generates the name for the cache (which will then be used to register the cache jmx bean): the name is by default set grails-cache-ehcache . When another second application or another application version is deployed registration will fail because the name already exists. The exception message is the following (indented for clarity): org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'ehCacheManagementService': Invocation of init method failed; nested exception is net.sf.ehcache.CacheException: javax.management.InstanceAlreadyExistsException: net.sf.ehcache:type=CacheM...

Create an OpenLayers map programmatically

Sometimes it is useful to abstract away the repetitive layer creation code with a configuration-based approach. For example consider this very simple map taken from the OpenLayers examples : How could we avoid repeating invoking the layer contructor and instead provde a framework that allows us to instantiate any layer with just configuration? The solution is quite simple.