Day 7 of the great Atlassian outage: IT giant still struggling to restore access

The good Atlassian outage is stumbling into a brand new week, with the biz reporting it has “rebuilt performance for over 35 p.c of the customers who’re impacted by the service outage,” which means the vast majority of these stay unable to entry their websites.

At this level it’s truthful to say the issue is extreme. It kicked off on 5 April and the IT large stated that whereas it was “working a upkeep script, a small variety of websites have been disabled unintentionally”.

Atlassian reported in a standing replace this morning that: “The rebuild stage is especially complicated resulting from a number of steps which can be required to validate websites and confirm information.” This will likely be of little consolation to the 65 p.c of affected clients questioning if signing up for Atlassian’s cloud was such a good suggestion in spite of everything.

We suspect that the “devoted workforce” Atlassian assigned to checking out the issue has but to take down the bunting from World Backup Day earlier than the incident occurred.

Jira Software program, Jira Work Administration, Jira Service Administration and Confluence are the largest segments nonetheless affected on the time of publication. Confluence is a web-based company wiki and Jira is extra about situation monitoring. Jira Work Administration is geared toward generic undertaking administration whereas Jira Service Administration turned up final yr as a part of a imaginative and prescient to show agile and DevOps ideas to the IT service desk.

The irony of the latter collapsing right into a heap resulting from a difficulty with a upkeep script is not going to have been misplaced on the affected customers.

Additionally nonetheless on the damaged checklist is IT incident-monitoring service, OpsGenie (acquired in 2018) and Atlassian’s Statuspage incident communication service – once more, the irony meter has gone off the size.

To be clear, solely a really small proportion of shoppers have been affected; on Friday we have been advised the determine was across the 400 mark. Nonetheless, we suppose being a type of 400 hasn’t precisely been enjoyable.

Atlassian has been down for six days. Our undertaking managers are climbing the partitions and our service request queues are all Slack threads.

I am presently doing the on-call rotation guidelines in a badly formatted Phrase doc somebody had managed to repeat from Confluence🙃

— Morten Linderud (@MortenLinderud) April 11, 2022

The Register contacted Atlassian to get extra data and an ETA for when the remainder of the info will likely be recovered. We’ll replace ought to the corporate reply. ®