74726278b9
core: Pause NackTimeout while in the plan_queue as progress is being made
2016-03-04 12:59:35 -08:00
c036fe6f19
Merge pull request #880 from achanda/docker_update
...
Test against latest stable docker
2016-03-04 10:52:37 -08:00
82d8210287
Remove documentation about assigning node-id
2016-03-04 10:36:30 -08:00
63b3703135
Always use the new package config if there is a conflict
2016-03-03 20:40:34 -08:00
bff2127ecf
Test against latest stable docker
2016-03-03 20:40:34 -08:00
1054c4cd0b
Merge pull request #879 from hashicorp/f-add-allocation-output-node-status
...
Include number of allocations in node-status
2016-03-03 15:37:00 -05:00
ce27b1e8ea
Add -allocs to help output
2016-03-03 14:46:24 -05:00
4e6737651a
Only fetch running allocs, use args, add docs
...
Only fetch running allocations, not total allocated ids, add
documentation, and cleanup things.
2016-03-03 14:09:07 -05:00
929e705ba1
Updating kill timeout adheres to operator specified maximum
2016-03-03 09:21:21 -08:00
c4a3f5047b
Include number of allocations in node-status
...
We recently ran into an issue on a small percentage of nomad-clients
where the nomad-client was running successfully, but due to a race
condition, could not correctly bind to the docker socket. This caused
all of our nomad jobs to be allocated to a single nomad-client instead
of being spread evenly across our clients. The only way to discover this
was to run `nomad node-status <node>` and count each job allocation per
node.
This can lead to a fairly long debugging process if there are several
nomad-clients. Including the number of allocations for each node in the
`node-status` command would save a large amount of debug time.
```
jake@biscuits [12:08:41] [~]
-> % nomad node-status
ID Datacenter Name Class Drain Status Allocations
2b0aabc5 dc1 biscuits <none> false ready 0
```
```
jake@biscuits [12:08:55] [~]
-> % nomad node-status
ID Datacenter Name Class Drain Status Allocations
2b0aabc5 dc1 biscuits <none> false ready 1
```
2016-03-03 12:19:56 -05:00
a44127aaa6
Merge pull request #768 from achanda/b-rkt
...
A number of rkt fixes
2016-03-02 21:29:48 -08:00
2bc6d8dee4
changelog
2016-03-02 17:00:40 -08:00
0b12e74a6e
Add /sbin and /usr/sbin/ to chroot
2016-03-02 16:32:31 -08:00
20746833cd
Merge pull request #874 from hashicorp/b-consul-concurrent-map-access
...
Remove a concurrent map access
2016-03-02 16:30:42 -08:00
f569f90267
Merge branch 'master' of github.com:hashicorp/nomad
2016-03-02 16:27:11 -08:00
144ccfb561
Killing a docker container that is dead is not an error
2016-03-02 16:27:01 -08:00
c2f83365b3
Merge branch 'master' of github.com:hashicorp/nomad
2016-03-02 15:04:21 -08:00
88107328c6
Updated changelog to contain libcontainer fix
2016-03-02 15:03:58 -08:00
2cefdff82b
Changelog
2016-03-02 15:00:53 -08:00
ef83066b7a
Fix location of volume mount
...
Also add a test for mounting alloc dirs
2016-03-02 14:04:14 -08:00
9e1055d595
Use NewWaitResult in the wait channel
2016-03-02 14:04:14 -08:00
5a3be40b8d
Set allocDir in handle
2016-03-02 14:04:14 -08:00
a78fc6707f
Run rkt interactively
2016-03-02 14:04:14 -08:00
29bcab9bfd
Cleanup if the plugin executor crashes.
2016-03-02 14:04:14 -08:00
8957726a0a
A number of cleanups
...
- Removed some unused variables.
2016-03-02 14:04:14 -08:00
a038d60a3b
Disable ipmasq in the default ptp network
2016-03-02 14:04:14 -08:00
77d9ba3526
Change name of the stdout file
2016-03-02 14:04:14 -08:00
fbc6f3f0b1
Do not hard-code binary location
2016-03-02 14:04:14 -08:00
d490df2bad
Use the plugin mechanism in the rkt driver
2016-03-02 14:04:14 -08:00
459ed48c43
Defer cleanup
2016-03-02 14:04:14 -08:00
e40c4630d7
Change how a result is returned form wait
2016-03-02 14:04:14 -08:00
579da2930a
Use less resources for containers
2016-03-02 14:04:14 -08:00
3da757addf
Print rkt and docker versions after installing
2016-03-02 14:04:14 -08:00
b47724e1e7
Put docker version in the script
2016-03-02 14:04:14 -08:00
0eac97b712
Add a shebang to docker update script
2016-03-02 14:04:14 -08:00
e1a1b7fc66
Enable running rkt tests in travis
...
- Introduces a travis target in Makefile to update docker and
install rkt
2016-03-02 14:04:14 -08:00
7719a1ce98
A number of rkt fixes
...
- Reorder rkt arguments to make it work
- Change a few things in test
2016-03-02 14:04:14 -08:00
b35e244c7d
Use a host volume to mount alloc and task dirs
...
host volumes are the preferred way to share data.
2016-03-02 14:04:14 -08:00
370332fdf3
Create the task and alloc dirs before proceeding
2016-03-02 14:04:14 -08:00
d7ebd7ab5e
Merge pull request #869 from hashicorp/f-alloc-id-env
...
Add Alloc ID/Name and Task Name to environment variables
2016-03-02 13:58:47 -08:00
e4c7c9f8b6
Remove a concurrent map access
2016-03-02 13:57:08 -08:00
6324e18cac
Merge pull request #855 from hashicorp/cgroups-pid
...
Updated logic for creation of cgroup
2016-03-02 12:57:25 -08:00
395d10eafc
Added a comment
2016-03-02 12:56:56 -08:00
04cb13f3a5
Added a test for making sure executor destroy works
2016-03-02 11:05:08 -08:00
ce798679fc
Merge pull request #870 from hashicorp/f-exec-driver-error-doc-update
...
Add a specific entry for the common error you may see
2016-03-02 10:01:22 -08:00
d34e5fe52f
Fix capitalization of Nomad in doc update
2016-03-01 23:52:42 -08:00
1374ba8101
Add a specific doc entry for the common error you may see when you are
...
using the exec driver and not running as root.
2016-03-01 23:51:11 -08:00
aaa23154aa
Not killing the plugin while destroying the cgroup
2016-03-01 17:51:54 -08:00
5748bd0516
Fixed the logic of cgroup creation
2016-03-01 17:01:45 -08:00
d7b425e5c1
Driver tests
2016-03-01 16:22:33 -08:00
Alex Dadgar