mirrors/kube-arangodb
1
0
Fork 0
mirror of https://github.com/arangodb/kube-arangodb.git synced 2024-12-14 11:57:37 +00:00
Code Activity

Make cleanoutserver step mandatory in drain documentation.

Browse source
This commit is contained in:
Max Neunhoeffer 2019-03-12 15:20:57 +01:00
parent 628f91385f
commit a5a7bf6680
No known key found for this signature in database
GPG key ID: 89A912AD5E343E1E
1 changed files with 21 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

28
docs/Manual/Deployment/Kubernetes/Drain.md
View file

@ -238,13 +238,14 @@ POST /_db/_system/_api/replication/clusterInventory
}
```
Check that for all collections the attribute `"allInSync"` has
the value `true`. Note that it is necessary to do this for all databases!
Check that for all collections the attributes `"isReady"` and `"allInSync"`
both have the value `true`. Note that it is necessary to do this for all
databases!
Here is a shell command which makes this check easy:
```bash
curl -k https://arangodb.9hoeffer.de:8529/_db/_system/_api/replication/clusterInventory --user root: | jq . | grep '"allInSync"' | sort | uniq -c
curl -k https://arangodb.9hoeffer.de:8529/_db/_system/_api/replication/clusterInventory --user root: | jq . | grep '"isReady"\|"allInSync"' | sort | uniq -c
```
If all these checks are performed and are okay, the cluster is ready to
@ -274,13 +275,14 @@ below, the procedure should also work without this.
Finally, one should **not run a rolling upgrade or restart operation**
at the time of a node drain.
## Clean out a DBserver manually (optional)
## Clean out a DBserver manually
In this step we clean out a _DBServer_ manually, before even issuing the
`kubectl drain` command. This step is optional, but can speed up things
considerably. Here is why:
`kubectl drain` command. Previously, we have denoted this step as optional,
but for safety reasons, we have made it mandatory now, since it is near
impossible to choose the grace period long enough in a reliable way.
If this step is not performed, we must choose
Furthermore, if this step is not performed, we must choose
the grace period long enough to avoid any risk, as explained in the
previous section. However, this has a disadvantage which has nothing to
do with ArangoDB: We have observed, that some k8s internal services like
@ -328,6 +330,12 @@ could use the body:
{"server":"PRMR-wbsq47rz"}
```
You can use this command line to achieve this:
```bash
curl -k https://arangodb.9hoeffer.de:8529/_admin/cluster/cleanOutServer --user root: -d '{"server":"PRMR-wbsq47rz"}'
```
The API call will return immediately with a body like this:
```JSON
@ -360,6 +368,12 @@ GET /_admin/cluster/queryAgencyJob?id=38029195
}
```
Use this command line to check progress:
```bash
curl -k https://arangodb.9hoeffer.de:8529/_admin/cluster/queryAgencyJob?id=38029195 --user root:
```
It indicates that the job is still ongoing (`"Pending"`). As soon as
the job has completed, the answer will be: