Page tree
Skip to end of metadata
Go to start of metadata

Weekly maintenance windows

In order to more quickly and regularly bring new Atlassian product features and fixes to you, we currently update your instances in weekly maintenance windows at the times listed below.

At some point during your maintenance window your Atlassian Cloud site will be unavailable for 15-30 minutes.

Your regionWindow

APAC (Asia Pacific)

Every Monday 11pm - Tuesday 3am Sydney time / Every Monday 12:00 - 16:00 UTC

EMEA (Europe, Middle East and Africa, including India)

Every Monday 1am - 5am Amsterdam time / Every Monday 00:00 - 04:00 UTC

AMER (America, including U.S.)Every Sunday 10pm - Monday 2am San Francisco time / Every Monday 06:00 - 10:00 UTC

Keep in mind that your local time might vary depending on your specific Daylight Savings Time schedule.

Getting notified about upcoming upgrades and news

We use the Atlassian Cloud blog to announce upcoming upgrades and the completion of them. We try to publish information about upcoming upgrades a few days prior to the weekly maintenance so that you can prepare for them if needed.

If you'd like to get notified about future upgrades and news, please subscribe to our blog.

Communication on maintenance windows

  • Regular weekly maintenance windows: We will not send maintenance notifications unless we need to deviate from the weekly schedule.
  • Other types of maintenance: For other required maintenance such as urgent fixes, we will notify you at least 12 hours in advance by sending an email message to all technical contacts listed under your account at my.atlassian.com, or by adding a banner to your site with information about the maintenance.

Changing your maintenance window region

The region of your maintenance window is determined by the time zone selected during the Atlassian Cloud signup process.

To change the window for your site maintenance, file a support ticket letting us know your preferred region. One of Support engineers will be in touch with you to update it.

 

  • No labels

27 Comments

  1. Hi

    Thanks for the advice.

    I am checking the meaning of the separate tables "Weekly maintenance windows as of March 16th" (all 4 hours) and "Regular weekly maintenance windows" (these are of different duration depending on timezone).

    Am I correct in understanding that the first table is general updates to common apps/infrastructure and the second table is specific to Atlassian Cloud user instances?

    Thanks

    1. Both tables refer to Atlassian Cloud user instances, the first refers to what the expiry will be from the 16th of March onward, whereas the second refers to the current window.

  2. What day of the week are the times listed in the "Weekly maintenance windows as of March 16th"?

    1. Apologies, that content was missing, it's still Sunday/Mondays. I've updated the table now.

  3. This page is not clear to me.

    1. What is the difference between:
      1. "Weekly maintenance windows as of March 16th"
        and:
      2.  "Regular weekly maintenance windows"

    2. What day of the week is the "Weekly maintenance windows as of March 16th"? It doesn't say here on this page.

    3. What is the impact to users? Will Atlassian Cloud products be unavailable during the maintenance periods? Or if not will performance be affected? 

     

    1. Sorry for the confusion Hugh, the first table refers to the windows after March 16 whereas the second table refers to the windows prior to March 16th.

      Maintenance windows are still Sunday/Monday, I've updated the table.

      Atlassian Cloud Applications will be unavailable for a short part (about 15-30 minutes) during those upgrade windows. We can't tell you exactly when during that 4 hour window they will be down, but they shouldn't be unavailable for the entirety of the window.

  4. Re: AMER (America, including U.S.) and Timezones in AEST.

    I would suppose most AMER people would assume AEST == EST. Which it's not.

    Perhaps better to declare times in ET and PT. 

    1. Martin,  AEST is Australian Eastern Standard Time.

    2. Yep, I changed the table to make timezones much more clear. The maintenance windows are locked to UTC though, so when daylight savings hits for various regions the window timing will change in your local time.

  5. What kind of disruption in service on Cloud based instances can we expect for your 4 hour maintenance window.  

    1. At some point during the 4 hour maintenance window your Atlassian Cloud instance will be down and unavailable for 15-30 minutes.

  6. This page is not clear.

    1. Apologies for that, I've updated it to be a little more clear, let me know if anything doesn't make sense now.

  7. This page looks unclear for me too.

    There is no information of UTC in the table "Weekly maintenance windows prior to March 16th".

    So I cannot interpret it to my local timezone.

    When do you carry out the periodic maintenance before March 16th in UTC?

    FYI, I live in Japan and my local timezone is JST (UTC+9 hours).

    Thanks.

    1. The maintenance window runs from 9pm - 11pm JST prior to March 16th. I've added UTC times to that table as well. (smile)

  8. Based on the comments above we know now that within the 4h window our cloud instance be be down at some stage between 15-30min.

    Some questions though:

    1. Are the REST APIs available during this time?
    2. Is there a particular REST API that we can query to check if our cloud instance is
      1. about to go down?
      2. currently down?

    I understand why you have a 4h window completely and I respect that you minimze the outage time to only 15-30min.

    However, we need to minimize the impact on business as well - currently the uncertainty of "15-30min at some stage within 4h" in effect means for us that we can't offer this QoS to our customers and therefore we are down for 4h in terms of business.

    Any tips how we can minimize this as close as possible to the actual 15-30min?

    Thank you.

    1. In answer to your questions:

      1. No, the REST API is unavailable during the 15-30 minute maintenance window.
      2. a. No, there's no endpoint for that.
        b. If you hit any REST endpoint during maintenance you'll get '503 Service Temporarily Unavailable', there is no particular endpoint for checking if it's down.

      As for mitigating the 15-30 minute impact, there isn't any way for us to tell you exactly when your instance will be down, unfortunately, as our system automatically runs through instances and upgrades them as resources become available, so the upgrade could happen for you at any time.

      If your business isn't 24 x 7 then perhaps another of the three windows is more convenient for you and your customers.

      We can't otherwise guarantee a specific time for your outage. If you want more guaranteed timeframes for maintenance then one of our Atlassian Expertsmight be a better solution for you. 

       

      1. Hi Matthew,

        Thanks for your answer - this is helpful.

        In regards to 2b): Are you saying that all REST APIs will go down with a 503 at exactly the same time? ... or will they go down one-by-one?
        We could use this error to check the "heartbeat" in regular intervals during the maintenance window and react to it on our end.

        We are indeed a 24x7 business, so changing time zones does not help us too much.

        Thank you.

        1. Each week not every application is upgraded (i.e. JIRA, Confluence and Bamboo), but if the application is upgraded then APIs for all upgraded products will fail at the same time. Each application takes a different amount of time to start up, but as it starts the APIs will become available. Generally, I would expect all APIs to become unavailable for 15 minutes and then become available again.

  9. Please, please stop putting a banner on every page with your announcements.  Use the login page please.

    Thanks!

  10. How can I stop seeing this announcement on every page?

  11. I'll add my vote to getting rid of the banner alerts on every page. Highly annoying.

  12. If you click the 'x' in the top right of the notification in both Confluence and JIRA the message should stay away unless you clear your cookies/cache or use a different browser. If you want the message removed from your site please log a support request at https://support.atlassian.com and we can remove the announcement for you.

  13. Actually in Safari the banner does not stay away. You can clear the banner in both JIRA & Confluence but whenever you open either back in a new browser tab or reopen the browser the banner comes back.

    1. The same in Chrome, and I'm not clearing cookies or cache.

    2. Also you cannot log a support request to get these turned off as according to Atlasssian support it's apparently it's not possible to do at this time with the JIRA cloud installation.

  14. How to find out which region my JIRA instance belongs to?