2,678 reputation
429
bio website comerford.cc
location Ireland
age 35
visits member for 2 years, 5 months
seen 1 min ago

I currently work for MongoDB, Inc. and have a background in operations, networking and system administration. You can find me in all the usual places, but probably the most useful and relevant are listed below:

Personal Site/Blog
LinkedIn
Twitter
Stack Overflow Careers


Apr
30
comment mongodb mystery crashes with sigkill under strace
Any chance that this is an Amazon VM?
Apr
25
comment Strange behaviour starting process with init script
In that case, the patch I mentioned above (which essentially stops the script from using numactl) should be a valid workaround for you until we fix the root issue with the bug. Just be aware that the init script may get overwritten if you upgrade at a later date.
Apr
24
comment Strange behaviour starting process with init script
ah, then numactl is installed (detected by the init script) but I believe the Linode instances are Xen VMs so I think you are OK. If in doubt, have a look here: serverfault.com/questions/471204/…
Apr
23
comment Strange behaviour starting process with init script
I think you are hitting the bug I added to the answer, looks like a numactl related issue. If you are on a non-NUMA machine, you could take a look at the patch here: jira.mongodb.org/browse/SERVER-8774 (for a NUMA enabled machine you should make sure mongod is run with numactl).
Apr
23
revised Strange behaviour starting process with init script
added a note about the relevant bug
Apr
23
answered Strange behaviour starting process with init script
Apr
22
answered Mongodb sharding - Init script for mongos process?
Apr
8
revised How to get the MongoDB' current working set size
Added note about 2.4 new feature
Mar
13
comment MongoDB default port change
you can also pass it as an argument to the binary in the upstart script as --port 9999 - exact same effect, but thought I would mention it as an alternative
Mar
8
answered Fresh MongoDB Using 100MB of Virtual Memory?
Mar
5
comment MongoDB: two nodes & primary election
actually, it does answer the question - 2/3 = majority, 2/2 = majority. The difference is that in a 3 node set if you lose a node you still have a fully functioning primary and a secondary, in a 2 node set if you lose a single node you are read only (no primary). You can run with 2 nodes, it just doesn't get you anything from a redundancy perspective - in fact you have made your system more likely to fail.
Feb
22
comment Mongodb keeps reporting “no primary found”, but shell works fine
Then I am going to guess that you have configured the driver incorrectly and that it can't connect to the primary at all. Is the client on a different host? If so, run a mongo shell from that host and try to connect to exactly what you have configured in the driver.
Feb
22
comment Mongodb keeps reporting “no primary found”, but shell works fine
it's not just sub-optimal, it's actually worse than having a single node, so if you simply turned off the arbiter and reconfigured to just a primary you would have a better set up than you do now
Feb
21
answered Mongodb keeps reporting “no primary found”, but shell works fine
Feb
20
comment How much of hardware resource is required for Graylog2?
then there's very little data in MongoDB in that system - it may still grow over time as you add data, so I was basically warning about that behavior. I can only speak to that part of it I'm afraid - I've never used either Graylog or Elasticsearch
Feb
15
comment mongo and app server cpu
Can you mention what version of MongoDB you are using?
Feb
14
answered How much of hardware resource is required for Graylog2?
Feb
4
awarded  Popular Question
Jan
25
comment MongoDB replicaSet failover incurs 1000x query response time increase
There are no slow queries being logged in the database in the second log output besides a getmore on the oplog (which is replication). That would suggest that the delay is on the PHP side rather than the DB, but you would need to add some debug to figure out where. Couple of quick checks - what happens if you just shut down the secondary rather than firewalling off and/or what if you actively reject (RST) rather than DROP - does the behavior change? I'm wondering if the detection of the new layout taking a long time might be an issue - failing more quickly would be a way to find out
Jan
24
awarded  Yearling