Jira Service Management 中断历史

1. 2026年9月3日 UTC 14:20investigating

   我们正在调查影响Jira服务管理的问题。 受影响的用户在观看或与受影响的服务领域互动时可能遇到错误或缓慢。 我们的小组正在积极调查根源。 我们将在下一个小时或更早提供下一次最新情况.

2. 2026年9月3日 UTC 15:20investigating

   我们查明了影响Jira服务管理的问题的根源,目前正在部署一个固定方案。 我们期待不久复苏,并将继续密切监测局势。 我们将在60分钟后提供下一次更新,如果有材料更新,则更早提供.

3. 2026年9月3日 UTC 16:15monitoring

   Jira Service Management的性能退化问题已经得到解决,目前所有受影响客户的服务都正常运行。 我们将继续密切监测执行情况,以确认稳定.

4. 2026年9月3日 UTC 17:03resolved

   2026年9月3日,一些Jira Service Management的用户可能会在网络上看到受影响的服务领域或性能退化时出错. 这个问题现已解决,这项服务对所有受影响的客户正常运行.

自动翻译自官方事件更新。

1. 2026年9月1日 UTC 10:11monitoring

   我们确定了影响Jira工作项目过渡的问题的根源,并采取了补救措施。 我们目前正在看到复苏的迹象,并将继续进行监测。 我们将在今后1小时内提供下一次最新情况,如果有实质性最新情况,则更早提供.

2. 2026年9月1日 UTC 11:44resolved

   2026年9月1日,从协调世界时04:22到协调世界时10:59,一些Jira用户可能经历了服务中断. 这个问题现已解决,服务正常向所有客户开放.

自动翻译自官方事件更新。

1. 2026年8月27日 UTC 04:22investigating

   我们的小组已查明这一问题也影响到Jira服务管理。 我们将尽快进一步更新.

2. 2026年8月27日 UTC 04:25investigating

   我们正在调查一个获取产品的问题,这些产品正在影响Conculation、Jira Cloud和Jira Serivce管理客户。 我们的小组正在紧急调查这一问题,如果有资料,我们将在一小时内或更早地提供进一步的细节.

3. 2026年8月27日 UTC 04:33identified

   我们的小组已经确定了解决影响Jira Cloud、Jira服务管理和协同的问题的可能办法。 目前正在受影响地区推广,我们将在一小时内更新,如果有进一步的信息,则将提前更新.

4. 2026年8月27日 UTC 04:51monitoring

   有关装载Jira Cloud、Jira Service Management和Conclus的问题目前正在恢复之中。 我们将继续密切监测,以确认稳定性,并在一个小时内提供最新情况,或者在获得更多信息后更快.

5. 2026年8月27日 UTC 05:16resolved

   在协调世界时2026年8月27日,包括Jira Cloud,Jira Service Management和相交公司在内的Atlassian产品出现中断,受影响用户无法获得服务. 这个问题现已解决,这项服务对所有受影响的客户正常运行。 我们完成对事件的全面调查后,将在本页公布这一事件的公开尸检结果.

6. 2026年9月7日 UTC 02:53postmortem

   页:1 2026年8月27日,协调世界时03:51至04:35,西欧和东南APAC地区的阿特拉斯斯客户无法到达阿特拉斯云产品,包括阿特拉斯云分析,康帕斯,相会,焦点,卫视,吉拉,JPD,JSM,Opsgenie,Rovo Search,和人才. 这次事件的起因是 我们代理舰队的路由层有缺陷的代码更改 这导致HTTP 404出错通过西欧和东南APAC区域的这些代理车队向所有客户流量服务. 我们的自动监控系统在1分钟内发现了这个事件,并通过回滚到之前的配置发布中来缓解,这让代理机队进入了已知的良好状态. 在代理机队恢复后,随着服务器机队适应流量的涌入,一些Jira和Congress客户(无论位置)在协调世界时04:33至04:53之间收到了HTTP 503出错. ** 缩略语** 总体影响分为两个独立的时期: - 首先,2026年8月27日,协调世界时03:51至04:35之间,关于阿特拉斯西亚分析、指南、聚合、焦点、警卫、吉拉、JPD、JSM、Opsgenie、Rovo搜索和人才。 该事件给来自西欧和东南APAC地区的客户流量造成服务中断,HTTP 404错误被退回. 不论来源地,对客户API的交通没有影响。 - 第二,2026年8月27日,协调世界时04:33至04:53之间,在HTTP 404出错解决后,一些Jira和Conculation客户收到了HTTP 503出错:在Ap-东南-2区~32%的请求,在eu-西-1区~2.5%的请求,在我们-西-2区和AP-东南-1区~2.5%的请求,在所有其余全球区域~1%的请求. [原件:英文] 这件事情的起因是 密码更改为我们的代理舰队之一 然而,所部署的代码有缺陷,而人工和自动测试没有发现。 这导致代理级的租户查询功能配置不当,导致流量没有有效的网络路径,导致HTTP 404出错通过西欧和东南APAC地区的这些代理车队向所有客户流量提供服务. 代理级影响恢复后,对一些Jira和Conculting客户造成了第二期影响。 由于代理机队向Jira和Concult服务减少,Jira和Concult服务级自动缩减了服务客户流量的服务器数量. 代理机队功能被恢复后,流量的猛增使Jira服务层不堪重负,它还原了HTTP 503s的高速率,直到再次自动缩放. ** 紧急行动计划和下一步** 我们知道停电会影响你的生产力。 虽然我们有一些测试和预防程序, 但这个具体问题并没有被确定, 因为改变与非常具体的边缘案例有关, 我们正优先采取以下改进行动来帮助避免此类事件再次发生: - ** 增加进一步的部署保障:** 受影响代用级别配置的部署管道使用遗留系统,没有正确配置适当的缓慢区域推出功能。 我们已经增加了正确的保障措施,我们正在将这些措施移至我们的连队标准部署管道,其中包括我们默认能够进行的所有安全控制。 我们还正在审计所有联网服务,以证实它们都在新的标准管道部署服务上。 - ** 加强服务队规模政策:** 我们的Jira和Confactive服务车队的自动缩放配置过于严格。 我们正彻底审查并更新所有复杂的Jira和相通汽车缩放政策, 我们向服务在这一事件中受到影响的客户道歉;我们正在立即采取步骤改善平台的性能和可用性。 谢谢 阿特拉斯客户支助

自动翻译自官方事件更新。

1. 2026年8月25日 UTC 08:42investigating

   我们正积极调查有关业绩退化的报告,这些报道导致对Jira、JSM和Conference的电子邮件、移动和不及时通知的延迟性增加。 我们的小组正在调查这一问题,并将在今后一小时内分享更多信息.

2. 2026年8月25日 UTC 09:44investigating

   我们的小组已经查明了受影响地区,并正在积极努力减轻这一问题。 我们将在一小时内提供进一步的最新情况,如果有重大进展要分享,则更早提供.

3. 2026年8月25日 UTC 10:49investigating

   我们的小组没有采取什么纠正措施来缓解一个受影响地区的问题,并且看到一些改进。 我们继续监测局势,并将采取必要行动,直至事件完全恢复。 我们将在今后两个小时内提供进一步的最新情况,或者,如果我们有重大进展可以分享,则更早地提供.

4. 2026年8月25日 UTC 13:01monitoring

   我们的小组继续缓解这一问题,并看到一些改进。 我们继续监测局势,并将采取必要行动,直至事件完全恢复。 我们将在今后两个小时内提供进一步的最新情况,或者,如果我们有重大进展可以分享,则更早地提供.

5. 2026年8月25日 UTC 14:33resolved

   2026年8月25日,一些用户在电子邮件,移动和不便通知中可能遭遇了更大的延缓. 这个问题现已解决,服务正常适用于所有客户.

自动翻译自官方事件更新。

1. 2026年8月24日 UTC 19:22investigating

   我们正积极调查有关业绩退化的报告,这些报告导致对JIRA、JSM和Confact的电子邮件、手机和Slack通知的延迟性增加。 我们将在此分享更多信息.

2. 2026年8月24日 UTC 20:25investigating

   我们的团队确定了三个受影响的地区,并继续调查影响JIRA、JSM和Confact的电子邮件、手机和Slack通知的日益延迟.

3. 2026年8月24日 UTC 21:30identified

   我们已查明了这一问题的可能原因,我们的小组正努力缓解这一问题。 受影响用户在电子邮件、手机和JIRA、JSM和Conference的Slack通知方面可能会遇到更多的不耐烦。 随着更多信息的提供,我们将继续在此分享更多的最新情况.

4. 2026年8月24日 UTC 23:15resolved

   2026年8月24日,一些JIRA,JSM,和相交用户可能经历了性能退化,导致电子邮件,手机和Slack通知的延迟性增加. 这一问题已针对新创建的通知得到解决,服务正常运行。 可能会有更早的通知正在处理之中,我们期望在接下来的几个小时内完成.

自动翻译自官方事件更新。

1. 2026年8月7日 UTC 11:30investigating

   We are investigating issues affecting your services. We are actively working to identify the cause and resolve the issue. We will provide further updates within one hour.

2. 2026年8月7日 UTC 12:22monitoring

   Some customers experienced issues accessing affected services in the Central European region. We have been able to apply the fix to mitigate the issues and we are seeing recovery. We continue to monitor the situation and should be fully recovered soon. We should be able to provide more updates in the next one hour or earlier.

3. 2026年8月7日 UTC 12:52resolved

   On August 7th - 2026, between 10:27 UTC to 11:49 UTC, some users in the Central European region may have experienced service access issues. The issue has now been resolved and the service is operating normally for all the customers.

1. 2026年8月7日 UTC 01:33investigating

   We are actively investigating reports of performance degradation affecting JIRA and JSM. Affected users may experience errors viewing issues and boards. We will share updates here as more information is available.

2. 2026年8月7日 UTC 01:39monitoring

   The performance degradation of JIRA has been resolved, and services are now operating normally for all affected customers. We'll continue to monitor performance closely to confirm stability.

3. 2026年8月7日 UTC 01:46resolved

   On August 7, 2026, JIRA, JWM, and JSM users may have experienced performance degradation when viewing issues and boards. The issue has now been resolved, and the service is operating normally for all affected customers.

1. 2026年7月21日 UTC 09:48identified

   我们正与我们的通知提供方合作解决这一问题。 我们将在24小时内或一旦获得更多信息后再提供最新情况.

2. 2026年7月22日 UTC 06:59identified

   Jira Service Management和Opsgenie客户使用基于印度的手机号码的少量语音通知可能失败. 客户可能看到声音调用日志失败,而不会收到预期的呼号。 我们的第三方通知提供者正在积极努力恢复语音通知发送,一旦问题得到解决,我们将提供最新情况。 如果您需要任何进一步的支持或信息,请联系我们的支持小组.

3. 2026年7月24日 UTC 13:07resolved

   我们的通知提供商已经查明了这个问题的根源,它正在给一些设在印度的手机号码造成语音呼叫故障。 我们理解这可能造成的不便 并欣赏你的耐心 当我们的供应商 努力恢复服务。 如果您在语音通知中遇到任何问题,请联系我们的支助处寻求援助.

自动翻译自官方事件更新。

1. 2026年7月20日 UTC 20:18investigating

   我们正在积极调查影响Jira和Jira服务管理自动化规则的业绩退化报告。 我们将在此分享更多信息.

2. 2026年7月20日 UTC 21:27investigating

   我们的小组正在继续调查对Jira和Jira服务管理自动化规则处决某些客户的拖延。 我们看到了一些改进的迹象,但受影响的用户仍然可能遇到延误。 我们在此分享最新消息.

3. 2026年7月21日 UTC 04:53identified

   团队已经确定了一个根源,在初步测试后,目前正在颁布缩放更改来帮助防止这些规则执行的拖延. 我们会在接下来的24小时内监控性能,并届时提供最新消息,如果能够确认改进到位,则会提前提供.

4. 2026年7月22日 UTC 00:56resolved

   我们的小组已经能够证实,所有规则的处决都是正常进行的,这个问题现在已经得到解决.

自动翻译自官方事件更新。

1. 2026年7月20日 UTC 16:03investigating

   我们正在积极调查影响Jira和Jira服务管理自动化规则的业绩退化报告。 我们将在此分享更多信息.

2. 2026年7月20日 UTC 16:58identified

   我们已查明了这一问题的可能原因,我们的小组正努力缓解这一问题。 但是,我们目前看到一些改善的迹象,受影响的用户在Jira和Jira Service管理自动化规则的处决方面可能会遇到延误。 我们将继续在这里分享更多的更新信息.

3. 2026年7月20日 UTC 17:23resolved

   2026年7月20日,一些Jira和Jira Service Management用户可能经历了自动化规则处决的性能退化. 这个问题现已解决,这项服务对所有受影响的客户正常运行.

自动翻译自官方事件更新。

1. 2026年7月6日 UTC 09:23investigating

   我们正在调查影响你服务的问题 我们正在积极努力查明原因并解决问题。 我们将在一个小时内提供进一步的最新资料.

2. 2026年7月6日 UTC 10:04monitoring

   我们成功地应用了修复方法,并正在看到恢复。 自动化能力正在得到恢复并应当正常运行。 由于无法提供服务,在停用期间提交的数据处理请求没有完成。 我们继续监测局势,并期望在下一个小时内全面恢复。 我们将在今后一个小时或更早地提供补充资料.

3. 2026年7月6日 UTC 11:46investigating

   我们仍在看到一些与欧盟区域用户搜索、用户选择和资产有关的问题。 我们继续调查这一局势,并将在今后一个小时或更早地提供更多的最新情况.

4. 2026年7月6日 UTC 12:25monitoring

   我们已运用这一办法来缓解这一问题并正在恢复。 我们继续监测局势,并应尽快完全恢复。 我们应该能够在下一个小时或更早时提供更多的最新情况.

5. 2026年7月6日 UTC 13:04resolved

   2026年7月6日 06:51 UTC至09:39 UTC部分用户可能经历过与自动化有关的问题,而08:10 UTC至11:50 UTC部分用户可能经历过与欧盟地区用户搜索,用户选择器和资产有关的问题. 这些问题现已得到解决,服务正常适用于所有客户.

6. 2026年7月17日 UTC 16:59postmortem

   □ 总结 在协调世界时2026年7月6日,6:51至20:17之间,欧洲(欧盟)地区使用云产品的Atlassian客户经历了影响自动化规则执行,用户搜索,用户取取用,及相关工作流程的服务中断. 事件开始于一个核心身份服务机构在欧盟地区经历了数据库饱和,时间不长和出错率上升. 在应急人员努力恢复能力的同时,采用了紧急缓解措施,阻碍自动化规则进入身份终点。 与此同时,身份延迟度的提高导致下游用户搜索服务出现连锁故障,使用户搜索和从多个产品中取取取用户的经验降低。 在增加了额外的数据库读取能力后,服务被逐步恢复,紧急区块被移除,用户搜索服务被手动提升. □ 影响 该事件在协调世界时6:51至20:17之间于2026年7月6日影响到欧盟地区多个Atlassian产品的客户. - ** 核心身份服务退化:** 在欧盟区域,产品间歇性准入率下降。 - ** 自动规则执行失败:** 自动化规则使用默认用户(Automation for Jira)在Jira,Jira Service Management,以及Jira产品发现在欧盟区域以权限错误失败. - ** 用户搜索和取用失败:** 在欧盟区域,用户搜索成功率大幅下降,产品中基本上没有取用用户的能力。 □ 根源 事件源于近期的配置改变,缩短了核心身份服务缓存寿命. 随着欧盟工作日流量于2026年7月6日猛增,更大比例的请求开始直接进入后援数据库,而不是从缓存中送达. 欧盟数据库没有足够的区域能力来吸收增加的负载,数据库CPU达到了饱和度,导致断断续续的访问率被否决。 作为缓解措施的一部分,采用了应急区块来减少饱和身份服务数据库的负荷并优先恢复核心身份服务。 这无意中阻止了自动化规则通过他们的执行前许可检查,导致其失败. 延长时间主要是由于处理限制节奏导致的二次自动化规则故障. 与此同时,持续退化导致依赖这些检查的产品间歇性增加和间歇性故障,并导致下游用户搜索服务出现连锁故障。 因此,在恢复权限服务以及扩大用户搜索服务以恢复能力之前,欧盟地区基本上无法获得用户搜索和取取取用户的经验。 □ 补救行动计划和下一步 我们知道停电会影响顾客的生产力。 Atlassian正在优先采取以下行动来帮助防止今后发生类似事件: - ** 改进高效紧急缓解措施的保障措施:** 需要更多审查Harden操作工具,以便采取减缓措施,使客户工作流程无法运作。 - ** 加强身份基础设施的能力并改变安全性:** 审查区域数据库的尺寸和容量,以减少在交通高峰期饱和的风险,并完善配置变化的评估程序,从而在生产推出前评估下游负荷影响。 - ** 增强抵御上游退化造成的连锁故障的能力:** 审计后压处理,断路器配置,以及服务中的自动升级行为,这样上游退化不会在用户搜索,取用者,以及相关经验中造成更广泛的产品影响. - ** 自动化规则故障的哈登重放机制:** 提高自动化规则的适应能力,以降低与依赖性事件有关的影响。 我们认识到,对我们的客户而言,关键可靠的产品工作流程是何等重要,我们向受这一事件影响的客户道歉。 谢谢 阿特拉斯客户支助

自动翻译自官方事件更新。

1. 2026年6月19日 UTC 22:46monitoring

   Jira和JSM的客户使用Microsoft OAuth自动创建从电子邮件发送的Jira门票无法读取电子邮件. 对所有新收到的电子邮件处理来说,这现已固定并运作。 事故期间受影响的客户需要重新授权微软邮件服务器恢复,下面是指令. 重新授权 Microsoft 邮件服务器 : 在 Jira 登录为管理员 选择 > 系统 选择邮件 > 来信邮件。 选择编辑收到的邮件服务器 。 选择重授权 。 输入您的微软证书以继续使用您的微软邮件服务器 .

2. 2026年6月26日 UTC 05:13resolved

   我们的团队主动联系了所有受影响组织,提供以下指示,重新授权其微软邮件服务器。 如果您仍在经历这个问题,请尝试按以下建议完成重新授权。 如果您认为与这起事件有关的进一步问题仍然存在,请联系阿特拉斯系统,寻求进一步援助.

自动翻译自官方事件更新。

1. 2026年6月10日 UTC 09:42investigating

   Users are facing degraded performance while accessing Assets page which is impacting Assets object creation, updation and deletion activities. Our teams are investigating actively and will be sharing more details in next 1 hour.

2. 2026年6月10日 UTC 10:00resolved

   On June 10 2026, between 08:10 and 09:08 UTC, users experienced degraded performance and inability to load the Assets page, which is impacting the ability to create, update, and delete assets. We have implemented a fix and services have been restored to normal operation for all affected customers.

1. 2026年6月10日 UTC 04:55investigating

   We are investigating reports of extended delays being experienced across tasks such as processing of Incoming Emails, Notifications relating to changes or creation of Work Items, in addition to delays in processing Scheduled Triggers from Marketplace apps. Our team is looking into this with urgency and we will provide further update within one hour as our investigation continues.

2. 2026年6月10日 UTC 05:39identified

   Our team has identified the root cause for this issue and has now deployed a fix that is rolling out across impacted products. At this time we are monitoring for the recovery of these events across Jira Service Management, Jira Cloud and Ecosystem app Scheduled Triggers. We will provide further update within one hour or when we have seen a full recovery of all task processing across these products.

3. 2026年6月10日 UTC 06:14monitoring

   Our team has now determined that performance has returned to expected levels for those impacted by this incident. Additional products that are now known to have been impacted by this incident are Confluence, Customer Service Management, and Jira Product Discovery. These products would have seen similar impact where email notifications or processing of events may have been delayed while this incident was underway. We are now monitoring for the full recovery of performance across all products and will update when we are confident this issue is fully resolved, or within one hour, whichever is sooner.

4. 2026年6月10日 UTC 06:31resolved

   We are now confident that any of the tasks that were impacted by this incident and were delayed should now be operating as expected. We apologize for any inconvenience that this issue may have caused.

1. 2026年6月9日 UTC 05:29investigating

   We are investigating reports of intermittent errors and degredated performance for some Jira Service Management and Jira Cloud customers. We will provide further updates within one hour.

2. 2026年6月9日 UTC 10:21identified

   Our team has now identified what we believe to be the root cause of this issue and is working on a fix. The expected fix for this issue will be rolling out progressively so customers should see resolutions to their issues viewing Jira boards and work items occurring over the next few hours. We will provide further update either when the update has fully completed or within 3 hours, whichever occurs earlier.

3. 2026年6月9日 UTC 10:36monitoring

   Our team has been able to apply the fix and mitigate the issue. The services have been restored to normal operations. We will continue to actively monitor over the next 8 hours to confirm stability.

4. 2026年6月9日 UTC 18:27resolved

   The issue has now been resolved, and the service is operating normally for all affected customers.

1. 2026年6月8日 UTC 01:53investigating

   We are actively investigating reports of performance degradation affecting Jira and Jira Software Management customers. We'll provide an update here within the next hour, or sooner if new information becomes available.

2. 2026年6月8日 UTC 02:36identified

   We have identified an issue causing degraded performance with Jira Software front-end experiences affecting view board, navigation, and backlog across multiple regions. Our engineering team is actively working on a resolution. We will provide an update within the next hour.

3. 2026年6月8日 UTC 02:56resolved

   On June 08 2026, between 01:30 and 02:40 UTC, users experienced degraded performance with Jira Software front-end experiences affecting view board, navigation, and backlog across multiple regions. We have implemented a fix and services have been restored to normal operation. We continue to monitor the situation closely.

1. 2026年6月2日 UTC 15:32investigating

   We are actively investigating reports of performance degradation affecting Jira and Jira Software Management customers. Some customers may experiences delays in rule execution. We'll provide an update here within the next hour, or sooner if new information becomes available.

2. 2026年6月2日 UTC 16:51resolved

   On June 2, 2026, some Jira and Jira Service Management users in the US-East region may have experienced performance degradation of automation rule executions. The issue has now been resolved, and the service is operating normally for all affected customers. Previously delayed automation rules are expected to be completed within 12 hours.

1. 2026年6月1日 UTC 08:06investigating

   We are investigating cases of partial-outage for Jira-software in the EU Region. We will provide more details within the next hour.

2. 2026年6月1日 UTC 08:06identified

   We have identified an issue causing degraded performance for Jira Software, Jira Work Management, and Jira Service Management in the EU region. affected users may experience 5XX errors when accessing issues, boards, and navigation. Our engineering team is actively working on a resolution. We will provide an update within the next hour.

3. 2026年6月1日 UTC 09:08monitoring

   We have implemented a fix for the issue causing degraded performance for Jira Software and Jira Service Management in the EU region. The impact has been mitigated, and we are actively monitoring the stability.

4. 2026年6月1日 UTC 13:23resolved

   This incident has been resolved. The issue causing degraded performance for Jira Software and Jira Service Management in the EU region has been fully mitigated. We have deployed a fix to prevent recurrence. If you continue to experience issues, please contact Atlassian Support. We apologise for the disruption and thank you for your patience.

5. 2026年6月12日 UTC 06:49postmortem

   **SUMMARY** On June 01, 2026, between 06:54 and 08:28 UTC, Atlassian customers using Jira, Jira Service Management, and Jira Work Management were unable to view issues, boards, and backlogs, and experienced errors when performing transitions, adding comments, and navigating the product. The event was triggered by caching infrastructure autoscaling failure. The issue only impacted customers hosted in Europe. The incident was detected within 8 minutes by our automated monitoring system and mitigated by progressively scaling the caching clusters manually. The total time to resolution was 1 hour and 34 minutes due to the progressive recovery. **IMPACT** The incident caused service disruption for customers hosted in the Europe region who were unable to access core experiences across Jira Software, Jira Service Management, and Jira Work Management. **ROOT CAUSE** The root cause of the incident was a capacity constraint on our caching infrastructure in one availability zone of our EU Central cloud region. As morning traffic increased, our caching clusters attempted to scale out automatically. However, the cloud infrastructure was unable to provision additional nodes in that availability zone due to insufficient underlying capacity. Our caching infrastructure is designed to scale across multiple availability zones to ensure balanced and resilient capacity. On this occasion, a capacity constraint in one availability zone prevented the scaling operation from completing as expected. With the caching layer unable to grow to meet demand, the clusters began to experience network and/or CPU saturation and could not handle all incoming requests within the expected timeout window. Since a single user request to Jira typically involves multiple cache calls, timeouts compound rapidly. As more threads became blocked waiting on timed-out or slow cache operations, the application nodes reached its busy thread limit and began shedding load. While additional application nodes were added, these also experienced the same thread limit saturation due to the saturation of the cache layer resulting in widespread 503 service unavailable errors for customers. **REMEDIAL ACTIONS PLAN & NEXT STEPS** We know that outages impact your productivity. In addition to existing safeguards and monitoring, Atlassian is prioritizing the following actions to help prevent similar incidents in future: - **Increased minimum caching cluster size**: We have increased the minimum number of nodes in all affected caching clusters in the EU Central region. This ensures sufficient capacity headroom to handle peak traffic without relying on autoscaling, reducing the risk of a similar failure occurring. - **Change autoscaling behaviour**: We are working on improvements to our autoscaling behaviour to fall back to other availability zones when capacity is unavailable in a single zone. We apologize to customers whose services were impacted during this incident; we are taking immediate steps to improve the platform’s performance and availability. Thanks, Atlassian Customer Support

1. 2026年5月14日 UTC 05:03investigating

   We are aware that the current issues with accessing Atlassian products are impacting additional products. We have now expanded the impact of the incident to cover the known impacted products at this time. We will provide further update within one hour or sooner as information is available.

2. 2026年5月14日 UTC 05:11identified

   Further confirmations of impacted products have now been added to the incident. Our team is investigating the issue with urgency and we will provide further update as soon as it becomes available.

3. 2026年5月14日 UTC 05:24identified

   It is likely if you are experiencing any issues relating to logging in or accessing Atlassian products at this time it is likely due to this ongoing incident. We are continuing to receive reports about expanded product impact resulting from this incident. While our team continues to investigate the issue with urgency, we will continue to provide further updates here with additional information. We will provide further update within one hour, or sooner as further information becomes available.

4. 2026年5月14日 UTC 05:34identified

   Our team has identified the root cause of this issue and is now actively working on mitigating the issue with accessing Atlassian products. At this time, Atlassian customers should also be able to once again raise support requests with our team. We will provide further update within an hour as we are able to progress mitigating this issue.

5. 2026年5月14日 UTC 05:43monitoring

   Our team has implemented a mitigation for this issue and we are now seeing recovery across Atlassian products. We will continue to monitor this issue for any ongoing concerns, and provide further updates here within an hour as we are able to confirm a full recovery has taken place.

6. 2026年5月14日 UTC 06:40monitoring

   We are now able to see recovery for all impacted products, and users should be able to access their products as expected. Our team is continuing to monitor all products and services to ensure there is no further impact, and we will provide further update when this has been validated and the incident is closed.

7. 2026年5月14日 UTC 06:41monitoring

   We are approaching full system recovery at this time, and are performing final confirmations that services are restored.

8. 2026年5月14日 UTC 06:48resolved

   All products and services impacted by this incident should now be fully recovered, and this incident is resolved.

9. 2026年5月19日 UTC 19:06postmortem

   ### Summary On May 14, 2026, between 04:30 and 05:26 UTC, Atlassian customers experienced widespread service disruption across multiple Atlassian Cloud products. The issue was caused by a race condition in our internal deployment orchestration platform during a routine rollback operation of a core identity service in the us-east region. This race condition resulted in insufficient capacity for the identity service in the affected region which started returning errors to dependent products. The incident was detected within a minute by automated monitoring systems and mitigated in 56 minutes. ### **IMPACT** During the incident, customers attempting to access Atlassian Cloud products in the us-east region experienced authentication and permission failures and were unable to access services. Customers also experienced errors when accessing the support portal until Atlassian fell back to an alternate support method. This was caused by a core identity service in the us-east region becoming unavailable. Affected products included Atlassian Administration, Atlassian Analytics, Bitbucket, Compass, Confluence, Jira, Jira Product Discovery, Jira Service Management and Trello. Some users outside us-east may have been affected in certain scenarios. ### **ROOT CAUSE** The incident was caused by a race condition in our internal deployment orchestration platform during a routine rollback operation of a core identity service in the us-east region. This race condition resulted in insufficient capacity for the identity service in the affected region which started returning errors to dependent products. ### **REMEDIAL ACTIONS PLAN & NEXT STEPS** We know that outages impact your productivity. Atlassian is prioritizing the following actions to help prevent similar incidents in future: * **Refine deployment orchestration safeguards** * Harden our deployment platform to prevent similar race conditions or resulting capacity loss during a rollback operation. * Streamline mitigation steps when a service becomes unavailable in a region. * **Reduce cross-region impact** * Improve regional isolation and fallback handling so an issue affecting a single region is less likely to impact customers or product functionality in other regions. We recognise how critical reliable access to Atlassian products is for our customers' productivity, and we apologize to customers who were impacted by this incident. Thanks, Atlassian

1. 2026年5月14日 UTC 01:15investigating

   We are aware that some users are experiencing an excessive number of notifications arriving from Jira Service Management (Assist) as our team replays delayed events experienced in this prior incident: https://jira-service-management.status.atlassian.com/incidents/v307yb60q5g7 At this time we expect that the vast majority of these notifications will be arriving within the next two hours. We will provide further update at that time or sooner if further information becomes available.

2. 2026年5月14日 UTC 03:52investigating

   Some of the users continue to experience an excessive number of notifications arriving from JSM (Assist) as our team replays delayed events experienced in this prior incident: https://jira-service-management.status.atlassian.com/incidents/v307yb60q5g7 At this time we expect that these notifications will be arriving for the next few hours, until the recovery of the delayed events are processed which is taking longer than expected (ETA approx. 4 hours). We will provide further updates within 6 hours or sooner if the recovery occurs.

3. 2026年5月14日 UTC 09:41resolved

   This incident has been resolved. During the incident window, some customers using Atlassian Assist experienced excessive notifications as our team replayed delayed events caused by this prior incident: https://jira-service-management.status.atlassian.com/incidents/v307yb60q5g7 Our engineering team identified and addressed the issue. All services have fully recovered, and notification delivery is operating normally. No data loss occurred as a result of this incident. We apologize for the disruption and any confusion caused by the excessive notifications.