Alexey Kovyrin's dashboard

Alexey Kovyrin asked a question in Cloudera on July 01, 2009 21:00:

Public repo for cloudera ec2 hadoop scripts

Do you have your scripts somewhere in a public git/svn/etc repo where we would be able to help with its development? Our company is going to use your scripts and we'll definitely going to patch them to be more useful/stable/etc.

Alexey Kovyrin started following the question "Why are logs for EC2 defaulting to the small root partition?" in Cloudera.

Alexey Kovyrin replied on July 01, 2009 20:53 to the problem "Cloudera's Distribution for Hadoop EC2 with EBS" in Cloudera:

I've managed to get through all the paces of bringing up the cluster and here is a quick hack I've used to get past this formatting problem:


[root@newconversions1 hadoop-ec2]# git show 4d02fa143d70e51309e1d923ab482113c67fc368
commit 4d02fa143d70e51309e1d923ab482113c67fc368
Author: root <root@newconversions1.scribd.com>
Date: Wed Jul 1 14:29:15 2009 -0500

Wait for storage before formatting it

diff --git a/hadoop/ec2/storage.py b/hadoop/ec2/storage.py
index 7dbf5a6..b03d9aa 100644
--- a/hadoop/ec2/storage.py
+++ b/hadoop/ec2/storage.py
@@ -64,7 +64,14 @@ def create_formatted_snapshot(cluster, size, availability_zone, image_id, key_na
print "Attaching volume to %s" % instance.id
volume.attach(instance.id, '/dev/sdj')

- run_command_on_instance(instance, ssh_options, 'mkfs.ext3 -F -m 0.5 /dev/sdj')
+ run_command_on_instance(instance, ssh_options, """
+ while :; do
+ echo 'Waiting for /dev/sdj...';
+ if test -e /dev/sdj; then break; fi;
+ sleep 1;
+ done;
+ mkfs.ext3 -F -m 0.5 /dev/sdj
+ """)

print "Detaching volume"
conn.detach_volume(volume.id, instance.id)



Alexey Kovyrin replied on May 07, 2009 20:02 to the question "rss" in Best Tech Videos:

I'm not sure what do you mean here, but we're going to release new version of the site where all the RSS/tags/catrgories stuff has been revamped and it'd be much easier to track what you really want.

Alexey Kovyrin started following the problem "uninitialized constant CommandError" in New Relic.

Alexey Kovyrin replied on December 25, 2008 16:13 to the question "Number of viewers not increasing at scribd" in Scribd (UNOFFICIAL):

Hi Nikhil,

This problem has been resolved now, sorry for the inconvenience.

Alexey Kovyrin started following the question "on passenger, will new relic wake up a sleeping site?" in New Relic.

kovyrin asked a question in New Relic on September 17, 2008 07:23:

Can I disable NR some way on per-box basis?

I have a set of hosts in our production cluster, but I do not want to buy NR for all of them since they all are identical. Can I disable NR some way on per-box basis? All boxes are deployed with the same environment - production - so it is not clear how could we deploy the stuff to all of them with only one box reporting to NR.