diff options
author | 2019-05-04 21:06:25 +0100 | |
---|---|---|
committer | 2019-05-04 16:06:25 -0400 | |
commit | 076b8d4fbae050f1f04475ad47709de872899374 (patch) | |
tree | 0dff3da9dbf1e6346babfda10e29749cb4514a92 /plugin | |
parent | 890cdb5cab8bb898dd048d853791ec82e28003ad (diff) | |
download | coredns-076b8d4fbae050f1f04475ad47709de872899374.tar.gz coredns-076b8d4fbae050f1f04475ad47709de872899374.tar.zst coredns-076b8d4fbae050f1f04475ad47709de872899374.zip |
plugin/health: add OnRestartFailed (#2812)
Add OnReStartFailed which makes the health plugin stay up if the
Corefile is corrupt and we revert to the previous version.
Also needs a fix for the channel handling
See #2659
Testing it will log the following when restarting with a corrupted
Corefile
~~~
2019-05-04T18:01:59.431Z [INFO] linux/amd64, go1.12.4,
CoreDNS-1.5.0
linux/amd64, go1.12.4,
[INFO] SIGUSR1: Reloading
[INFO] Reloading
[ERROR] Restart failed: Corefile:5 - Error during parsing: Unknown directive 'bdhfhdhj'
[ERROR] SIGUSR1: starting with listener file descriptors: Corefile:5 - Error during parsing: Unknown directive 'bdhfhdhj'
~~~
After which the curl still works.
This also needed a change to reset the channel used for the metrics
go-routine which gets closed on shutdown, otherwise you'll see:
~~~
^C[INFO] SIGINT: Shutting down
panic: close of closed channel
goroutine 90 [running]:
github.com/coredns/coredns/plugin/health.(*health).OnFinalShutdown(0xc000089bc0, 0xc000063d88, 0x4afe6d)
~~~
Signed-off-by: Miek Gieben <miek@miek.nl>
Diffstat (limited to 'plugin')
-rw-r--r-- | plugin/health/README.md | 7 | ||||
-rw-r--r-- | plugin/health/health.go | 3 | ||||
-rw-r--r-- | plugin/health/setup.go | 3 |
3 files changed, 3 insertions, 10 deletions
diff --git a/plugin/health/README.md b/plugin/health/README.md index a62a55e66..eb86635ad 100644 --- a/plugin/health/README.md +++ b/plugin/health/README.md @@ -76,10 +76,3 @@ Set a lameduck duration of 1 second: } } ~~~ - -## Bugs - -When reloading, the health handler is stopped before the new server instance is started. If that -new server fails to start, then the initial server instance is still available and DNS queries still -served, but health handler stays down. Health will not reply HTTP request until a successful reload -or a complete restart of CoreDNS. diff --git a/plugin/health/health.go b/plugin/health/health.go index eef45d2cf..55ff68407 100644 --- a/plugin/health/health.go +++ b/plugin/health/health.go @@ -28,6 +28,7 @@ func (h *health) OnStartup() error { if h.Addr == "" { h.Addr = ":8080" } + h.stop = make(chan bool) ln, err := net.Listen("tcp", h.Addr) if err != nil { @@ -51,8 +52,6 @@ func (h *health) OnStartup() error { return nil } -func (h *health) OnRestart() error { return h.OnFinalShutdown() } - func (h *health) OnFinalShutdown() error { if !h.nlSetup { return nil diff --git a/plugin/health/setup.go b/plugin/health/setup.go index 0e51ec5ea..11d498b09 100644 --- a/plugin/health/setup.go +++ b/plugin/health/setup.go @@ -32,8 +32,9 @@ func setup(c *caddy.Controller) error { }) c.OnStartup(h.OnStartup) - c.OnRestart(h.OnRestart) + c.OnRestart(h.OnFinalShutdown) c.OnFinalShutdown(h.OnFinalShutdown) + c.OnRestartFailed(h.OnStartup) // Don't do AddPlugin, as health is not *really* a plugin just a separate webserver running. return nil |