Thomas Orozco
ee8701aa97
Netlink: aggressively close cgroup fds
...
Not closing the FDs manually means we have to rely on garbage collection
to run before cgroup FDs are closed. If the system is running a lot of
load probes at a high-frequency (i.e. dynamic housekeeping isn't backing
off because of load variations), we can end up hitting our FD limit due
to keeping around lots of (useless) FDs.
2016-04-13 11:30:48 +02:00
Lei Xue
dbbe38dfed
re-order the import package
2015-11-30 16:43:22 +08:00
Victor Marmol
11462d80bc
Lowering log levels.
...
Reduce common logging using Kubernetes logging standards.
2015-04-13 15:05:41 -07:00
Victor Marmol
54bc33dd2c
Lowering log level for frequent events.
...
Lowering all frequent normal logs to v=3. Kubelet runs by default on
debug of v=2 and we don't want to log these events in that case.
2015-03-09 14:53:53 -07:00
Rohit Jnagal
872546ba3a
Bulk move current info api to info/v1. Making room for info/v2.
2015-03-04 00:47:28 +00:00
Rohit Jnagal
0b7e41c991
Fix typo in netlink example.
2015-01-30 16:56:34 +00:00
Rohit Jnagal
67af1a1bbd
Add scheddebug based cpuload reader.
...
Prefer netlink based reader when available as it provides stats for
processes in state other than running (sleep, diskio, uninterruptible).
2015-01-26 18:46:29 +00:00
Rohit Jnagal
1375f451b2
Refactor netlink implementation.
...
This allows us to plug in a scheddebug based interface.
2015-01-23 23:07:50 +00:00
