Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Answered
Is There Any Canonical Way To Migrate All Data Between Different

Is there any canonical way to migrate all data between different clearml-server s? Some time ago we were using trains server with quite a few valuable experiments on it. The idea of upgrading it in-place seemed too risky, so we decided to setup the new version alongside the old one instead. I think this is a typical scenario and should be covered by documentation. I know there is a guide for upgrading trains to clearml , but there seems to be no guide to my case

We ended up manually copying mongo database, preserving all old _ids, then copying all experiments cache. Now it sees the experiments from the web UI, but clicking on any of them results in "Fetch failed" exception

  
  
Posted 3 years ago
Votes Newest

Answers 10


ok, thanks, I'll give it a try

  
  
Posted 3 years ago

any idea what is missing?

  
  
Posted 3 years ago

What exactly we need to copy? I believe we have already copied everything, but it keeps throwing "Fetch experiment failed" error

  
  
Posted 3 years ago

SuccessfulKoala55 Turns out we have copied elasticsearch database as well. Also it seems that the error is thrown only for experiments with artifacts

  
  
Posted 3 years ago

Well, the best way is to simply copy the data folders when setting up your new server - this way you won't risk the old server and can get a new upgraded server with the existing data

  
  
Posted 3 years ago

image

  
  
Posted 3 years ago

Or maybe there is a log with something more informative which I could check up

  
  
Posted 3 years ago

For experiments with no artifacts, everything seems to work properly

  
  
Posted 3 years ago

Perhaps artifacts were uploaded to the fileserver? part of the backup/restore is copying the fileserver folder...

  
  
Posted 3 years ago

The docs contain backup and restore instructions, which is basically what you should do. The process includes copying all data folder, primerily for mongo and elastic

  
  
Posted 3 years ago
1K Views
10 Answers
3 years ago
7 months ago
Tags