181 reputation
110
bio website
location New York, United States
age 52
visits member for 4 years, 2 months
seen Jul 4 at 13:10

I've an eclectic background as a developer and a history of wearing many hats.

I currently work as an IT consultant typically working on integration projects for 'Wall Street' firms. I'm also the firm's in-house web developer and server admin, responsible for our internal web applications.

You can contact me through my Google Plus Profile - look for the 'Send an Email' link on the lower right hand side of the main panel.


1d
awarded  Notable Question
Jul
2
awarded  Curious
Dec
17
awarded  Popular Question
Feb
25
awarded  Popular Question
Jun
4
accepted MySql Data Loss - post mortem analysis - RackSpace Cloud Server
Jun
1
awarded  Teacher
Jun
1
revised MySql Data Loss - post mortem analysis - RackSpace Cloud Server
added 1 characters in body
Jun
1
revised MySql Data Loss - post mortem analysis - RackSpace Cloud Server
added 5 characters in body
Jun
1
revised MySql Data Loss - post mortem analysis - RackSpace Cloud Server
point to my answer for explanation
Jun
1
answered MySql Data Loss - post mortem analysis - RackSpace Cloud Server
May
31
comment MySql Data Loss - post mortem analysis - RackSpace Cloud Server
Does that mean that if there were less than 100 MB in the buffer pool it should be written (flushed) within 10 seconds? Or that every 10 seconds it checks to see if there's 100 MB or more before executing a flush? We might not see 100 MB of new data (insert/update) for days. If it's the latter case, then it's consistant with the evidence. Otherwise, not so much.
May
31
comment MySql Data Loss - post mortem analysis - RackSpace Cloud Server
The mysql docs are confusing about the default InnoDB flush method. Given "InnoDB ... does not use O_DSYNC by default ..." and "... InnoDB ... the default value or O_DSYNC ..." If the server was set to cache to the buffer pool only (O_DIRECT ?) what controls the flush interval? Cache size, time since last flush, or some combination? The server has very low volume use, so if it's cache size only, it might makes sense. I would have thought it would default to something like an hourly flush (at most) though.
May
31
revised MySql Data Loss - post mortem analysis - RackSpace Cloud Server
update details
May
31
asked MySql Data Loss - post mortem analysis - RackSpace Cloud Server
May
17
revised Monit is restarting apache2, should I fix the monit config, or the server config?
located an article that provides a good answer to the question - added as a supplement
Feb
17
awarded  Commentator
Feb
17
comment Monit is restarting apache2, should I fix the monit config, or the server config?
Yes, bumping the memory worked. Then we added more vhosts to the server and then bumped the machine size.
Sep
7
awarded  Citizen Patrol
Apr
26
accepted Monit is restarting apache2, should I fix the monit config, or the server config?
Apr
26
comment Monit is restarting apache2, should I fix the monit config, or the server config?
Thanks. Just did some testing, watching top while someone hit the new wordpress site, and saw the apache2 processes jump up to RES 45 to 48 -- and stay there (ie. not obviously leaking). While free -m still reports 220 M in the -/+ buffers/cache line. So I moved the monit test to 250M -- and we'll see what happens. There is nothing else on the VPS other than apache2 serving those sites.