We currently use a very complicated setup for github/jira. There are 2 github repos with issues, and 3 Jira projects to track our work. We rely on a data synchronization mechanism to sync state and updates from github into jira.
Currently, the sync is almost never accurate. Some changes happen in Jira and are not synced to github, and some changes happen in github and are not synced to jira. It is virtually impossible to rely on either as an accurate source of truth for the state of our progress at any given time.
We should fix this. I’m not sure if that means that we fix the sync to perform better, or to simplify the architecture so we don’t need as much syncing in the first place..
Possible Solutions
collapse number of repos/projects where we create issues. Maybe we could create a single github repository for our issues and sync this repo to 1 Jira project. This reduces the complexity of syncing as there are fewer repos to look at and fewer projects to sync to/from
Remove one of the ticketing systems. We could say we only use Jira for tickets, or we only use github. There are reasons to say we should use either (github for community exposure and corporate consistency/visibility for jira) but we could decide on using just one system and then there is nothing to sync.
Use a different or better syncing mechanism?
what do we use now? I don’t even know? There are a few vendors available?
we could make the sync simpler. instead of using labels and status to sync state we could just use the sync to synchronize data and use Jira to update statuses..
To add the list of known problems with our current sync (off the top of my head - there are probably more):
changes or comments in jira aren't reflected in github - but jira tickets aren't "read-only" in any way to prevent it
milestones in github don't connect to anything, even though we use them in the same was sprints, and name them similarly.
closing a github issue doesn't close the jira ticket
tickets filed in jira aren't shown as created by the account in github (e.g. this github issue shows as being created by data-sync-user), but jira doesn't prevent tickets from being created in jira on the sync'd repos.
Problem
We currently use a very complicated setup for github/jira. There are 2 github repos with issues, and 3 Jira projects to track our work. We rely on a data synchronization mechanism to sync state and updates from github into jira.
Currently, the sync is almost never accurate. Some changes happen in Jira and are not synced to github, and some changes happen in github and are not synced to jira. It is virtually impossible to rely on either as an accurate source of truth for the state of our progress at any given time.
We should fix this. I’m not sure if that means that we fix the sync to perform better, or to simplify the architecture so we don’t need as much syncing in the first place..
Possible Solutions
┆Issue is synchronized with this Jira Task