Reputation
1,045
Top tag
Next privilege 2,000 Rep.
Edit questions and answers
Badges
8 21
Newest
 Caucus
Impact
~86k people reached

Aug
4
comment Fastest/scalable way to serve static files?
@user658911 let us continue this discussion in chat
Aug
4
comment Fastest/scalable way to serve static files?
Buffer-cache automatically uses RAM that is otherwise free. To increase it, add more RAM or reduce RAM usage by other things, but I'm not convinced that's really your bottleneck. Once this gets moved to Serverfault, it'll be seen by a crowd that more routinely deals with this stuff.
Aug
4
comment Fastest/scalable way to serve static files?
This is a question for Serverfault. BTW: Its surprising you're seeing performance bottlenecks serving static files (are you really? and why is it reading from the hard disk each time, have you disabled the buffer cache?)
Jul
13
comment Bind9 Ubuntu and Google DNS
504 errors come from your web server, not your DNS server. But this belongs on Serverfault.
Jun
8
comment introduce latency between ec2 instances
This probably belongs on Serverfault, but the normal way on Linux would be with the network emulator queuing discipline. linuxfoundation.org/collaborate/workgroups/networking/netem
May
10
comment ssl communication handshake question
Yes, its configurable. See httpd.apache.org/docs/2.0/mod/… for example.
May
9
comment Making SATA disk write cache safe
Quite possible. Soon I hope to pull that machine, and do extensive testing on it.
May
6
comment Making SATA disk write cache safe
Ummm, these are Seagate Constellation ES drives. The other server (the one with the 3ware card) has VeliciRaptor drives. No cheap desktop drives in sight. Though, honestly, I've got other machines with cheap desktop drives, they've proved only a little less reliable.
May
5
comment Making SATA disk write cache safe
Yeah. I have a 3ware one on a different box. The data is indeed safe, in the sense that had Sony bought some of those, PSN data would be safe from crackers. Every time a single drive timed out, it'd discard cache on all drives, leading to massive corruption. I turned off the cache.
May
5
comment Making SATA disk write cache safe
ST3500514NS are most definitely not SSDs. But that is interesting.
Mar
4
comment Experience with Intel X25-M 160GB and Oracle
@sciurus: When the write range is only 8GB, it can use the other 152GB as a scratch area, to avoid flash erases. When the whole disk is being hit, the -M has I believe a 7% or so scratch area, so there is a lot more overhead for flash erases & wear leveling. Also probably a harder hit on the block reallocation in the disk.
Mar
4
comment Experience with Intel X25-M 160GB and Oracle
@Kyle Brandt: Intel gives a write endurance w/ random writes of only 15TB, so that is indeed a concern. We'd be replacing them fairly often, it sounds like :-( Please make an answer out of that...
Mar
4
comment Experience with Intel X25-M 160GB and Oracle
@sciurus: Intel only claims 300 random writes per second for the X25-M. See intel.com/design/flash/NAND/mainstream/pdf/322697.pdf
Mar
3
comment Experience with Intel X25-M 160GB and Oracle
@Kyle Brandt: Curious about why you'd expect it to cause downtime, especially in RAID10? I don't see anything in the other question to suggest that.
Jan
21
comment Configuring RADIUS or LDAP on Supermicro (ATEN) IPMI
@Hunta: updated. Not sure if you're watching this. I think I've now got a full answer.
Jan
20
comment Configuring RADIUS or LDAP on Supermicro (ATEN) IPMI
@Hunta: I now have an answer from SuperMicro. Edited in.
Jan
19
comment How can I find automatically running scripts on a sever?
@Michael Mior: At least here, /etc/crontab actually runs /etc/cron.FOO.
Jan
4
comment Same netmask or /32 for secondary IP on Linux
Well, "wrong order by mistake" includes "administrator's mistake". The order IPs come up in from the config is deterministic enough (though things like parallel booting make it less so). /etc/network/interfaces is not sequential anymore, its normally the order udev adds the interfaces.
Dec
13
comment How to achieve zero down time
This belongs on Serverfault, but keep in mind that as downtime permitted approaches 0, cost approaches ∞.
Aug
13
comment Configuring RADIUS or LDAP on Supermicro (ATEN) IPMI
@aniel Lawson: 2.01 now. Turns out there are definitely some weird attributes that need to be set, they sent me a doc with some examples but no explanation. Still following up to find out what they mean...