Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Do not use versions to optimize cluster state copying for a first update from a new master #6466

Conversation

Projects
None yet
3 participants
@bleskes
Copy link
Member

bleskes commented Jun 11, 2014

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

NOTE: this a PR for the feature/improve_zen branch

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.
@@ -560,6 +558,10 @@ public static Client client() {
return client;
}

public static Client client(@Nullable String node) {
return node == null? client() : cluster().client(node);

This comment has been minimized.

Copy link
@martijnvg

martijnvg Jun 11, 2014

Member

I think node can't be null here.

This comment has been minimized.

Copy link
@martijnvg

martijnvg Jun 11, 2014

Member

argh... didn't read it properly... nevermind

@martijnvg

This comment has been minimized.

Copy link
Member

martijnvg commented Jun 11, 2014

LGTM

bleskes added a commit that referenced this pull request Jun 12, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466
@bleskes

This comment has been minimized.

Copy link
Member Author

bleskes commented Jun 12, 2014

Pushed in with a2ca26e

@bleskes bleskes closed this Jun 12, 2014

@bleskes bleskes deleted the bleskes:always_update_cs_from_master branch Jun 12, 2014

bleskes added a commit that referenced this pull request Jun 13, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jun 17, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jun 21, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jun 23, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jun 26, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 2, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 3, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 3, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 7, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 7, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 8, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 9, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 11, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 12, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit to bleskes/elasticsearch that referenced this pull request Jul 14, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes elastic#6466

bleskes added a commit that referenced this pull request Jul 16, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 17, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 21, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Jul 22, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Aug 4, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Aug 5, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Aug 14, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Aug 22, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Aug 23, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit that referenced this pull request Aug 27, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

bleskes added a commit to bleskes/elasticsearch that referenced this pull request Sep 1, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes elastic#6466

@bleskes bleskes added v1.4.0 labels Sep 2, 2014

bleskes added a commit that referenced this pull request Sep 8, 2014

[Discovery] do not use versions to optimize cluster state copying for…
… a first update from a new master

We have an optimization which compares routing/meta data version of cluster states and tries to reuse the current object if the versions are equal. This can cause rare failures during recovery from a minimum_master_node breach when using the "new light rejoin" mechanism and simulated network disconnects. This happens where the current master updates it's state, doesn't manage to broadcast it to other nodes due to the disconnect and then steps down. The new master will start with a previous version and continue to update it. When the old master rejoins, the versions of it's state can equal but the content is different.

Also improved DiscoveryWithNetworkFailuresTests to simulate this failure (and other improvements)

Closes #6466

@clintongormley clintongormley changed the title [Discovery] do not use versions to optimize cluster state copying for a first update from a new master Resiliency: Do not use versions to optimize cluster state copying for a first update from a new master Sep 8, 2014

@clintongormley clintongormley changed the title Resiliency: Do not use versions to optimize cluster state copying for a first update from a new master Do not use versions to optimize cluster state copying for a first update from a new master Jun 7, 2015

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.