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

add start-es-inserts-after-block option to elasticsearch plugin #1455

Closed
1 task
oxarbitrage opened this issue Nov 27, 2018 · 16 comments
Closed
1 task

add start-es-inserts-after-block option to elasticsearch plugin #1455

oxarbitrage opened this issue Nov 27, 2018 · 16 comments
Assignees
Labels
1b User Story The User Story details a requirement. It may reference a parent Epic. It may reference child Task(s) 6 Plugin Impact flag identifying at least one plugin

Comments

@oxarbitrage
Copy link
Member

oxarbitrage commented Nov 27, 2018

Due to the blockchain halt we recently had all ES nodes stopped working and replay is required.

This takes a huge amount of time. Most of the data is already on ES so i am thinking on an option start-es-inserts-after-block or similar that will start the heavy work after a block number passed, resulting in a faster synchronization.

As some important business rely on the ES plugin for their business, i propose that feature or something to speed up the process to be added.

Looking for your opinions.

Tasks:

  • start-es-after-block
  • [ ] dump-es-at-block
@crypto-1
Copy link

This feature would be very useful for CryptoBridge.

@clockworkgr
Copy link
Member

clockworkgr commented Nov 27, 2018

yaaaaaaaaaaaaaaaaaaay!

I pretty much suggested the same thing earlier today in TG Node Admins...

@cogutvalera
Copy link
Member

Very nice idea !

@sschiessl-bcp
Copy link

sschiessl-bcp commented Nov 28, 2018

Could an ES dump be provided as well that matches es-inserts-after-block=32600000?

That would be asweome!

@oxarbitrage
Copy link
Member Author

Could an ES dump be provided as well that matches es-inserts-after-block=32600000?

This is also a good idea. I added 2 tasks to this issue to do start-es-at-block and dump-es-at-block

@oxarbitrage oxarbitrage added 1b User Story The User Story details a requirement. It may reference a parent Epic. It may reference child Task(s) 6 Plugin Impact flag identifying at least one plugin labels Nov 28, 2018
@oxarbitrage oxarbitrage self-assigned this Nov 28, 2018
@sschiessl-bcp
Copy link

I suppose if the backend creates the dump, it could also read it, i.e. 'import-es-on-block'

Just to say it, intent was to use the native dump and import es functional

@sschiessl-bcp
Copy link

Maybe replace 'at' with 'before' or 'after', so the timeline is clear with respect to the block

@clockworkgr
Copy link
Member

i see no need to add this to the plugin...just documentation of how to use native ES dump/import

And maybe we can provide weekly ES snapshots (those of us running a node)

@sschiessl-bcp
Copy link

No need maybe, but it's certainly nice. Install ES in default, rest is done by plugin.

@clockworkgr
Copy link
Member

where would the plugin get the dump from?

@sschiessl-bcp
Copy link

sschiessl-bcp commented Nov 28, 2018

That command would have a file, or github reference to e.g. BitShares/bitshares-core-dumps as parameter

@oxarbitrage
Copy link
Member Author

  • i removed the dump option from the task list, lets keep it simple. i agree with @clockworkgr; it is probably better to use ES native tools or third parties(https://github.com/taskrabbit/elasticsearch-dump) to do this because also if we export we will want to import and that can start to became more complicated without need.
  • changed the option to start-es-after-block

@Agorise
Copy link

Agorise commented Nov 28, 2018

As we are bringing entire cities onto point-of-sale now, any down time could cost merchants dearly, so if this will help get nodes up faster, I'm all for it.

@oxarbitrage
Copy link
Member Author

Closed by #1455

Now ES nodes can recover from crashes in minutes by start saving data only after a specific block pass.

@clockworkgr
Copy link
Member

Closed by #1455

Now ES nodes can recover from crashes in minutes by start saving data only after a specific block pass.

closed by #1458... 1455 is this issue

@oxarbitrage
Copy link
Member Author

oops, sorry and thanks.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
1b User Story The User Story details a requirement. It may reference a parent Epic. It may reference child Task(s) 6 Plugin Impact flag identifying at least one plugin
Projects
None yet
Development

No branches or pull requests

7 participants