# I can only rebuild with SSH? **URL:** \ **Category:** Build Environment\ **Created:** [November 8, 2017, 8:37am UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932 "2017-11-08T08:37:22Z")\ **Posts on this page:** 13\ **Page:** 1
**Author:** ![FanaHOVA](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/fanahova/32/11359_2.png) [@FanaHOVA](https://discuss.circleci.com/u/FanaHOVA)\ **Post date:** [November 8, 2017, 8:37am UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/1 "2017-11-08T08:37:22Z")
Hi everybody! For the past couple of weeks it seems like my builds don’t have a rebuild option anymore, but only prompt me to rebuild with SSH. This happens for both failed and succeeded builds. I was trying to debug some flaky specs and needed to retry succeeded builds but couldn’t find an option to do so neither on the build page nor on the repo index since it only appears for failed ones. I ended up resolving it by using rerun on the workflow rather than just the tests but it’s obviously not ideal. Let me know if there’s some setting I need to change to get the old behavior back. ---
**Author:** ![liene](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/liene/32/14373_2.png) [@liene](https://discuss.circleci.com/u/liene)\ **Post date:** [December 5, 2017, 12:23am UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/2 "2017-12-05T00:23:42Z")
Hi @FanaHOVA! That’s correct: plain rebuilds aren’t currently available on builds that are part of a Workflow. (See this post for a bit more detail: [https://discuss.circleci.com/t/rebuild-workflow-button-in-build/17807/2](https://discuss.circleci.com/t/rebuild-workflow-button-in-build/17807/2)) The salient part is this: > We have plans to add “Rerun Workflow” button on the jobs (build) page. For now, you would have to navigate to the “Workflows” page to rerun a Workflow or rerun a Workflow from failed jobs. ---
**Author:** ![FanaHOVA](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/fanahova/32/11359_2.png) [@FanaHOVA](https://discuss.circleci.com/u/FanaHOVA)\ **Post date:** [December 5, 2017, 8:11am UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/3 "2017-12-05T08:11:45Z")
@liene yea that’s what I ended up doing, thanks for following up! ---
**Author:** ![honnix](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/honnix/32/17875_2.png) [@honnix](https://discuss.circleci.com/u/honnix)\ **Post date:** [February 21, 2018, 11:16pm UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/4 "2018-02-21T23:16:19Z")
Hi @liene , Has this feature been implemented? This [doc](https://circleci.com/docs/2.0/ssh-access-jobs/#nav-button) says it’s there, but I still don’t see it for my failed job. Thanks. /H ---
**Author:** ![iwasrobbed](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/iwasrobbed/32/18865_2.png) [@iwasrobbed](https://discuss.circleci.com/u/iwasrobbed)\ **Post date:** [May 18, 2018, 7:02pm UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/5 "2018-05-18T19:02:26Z")
You should be able to re-run only the failed or SSH jobs via the workflows tab ![59%20PM](https://global.discourse-cdn.com/circleci/original/3X/6/4/64049616e7aeb01d369e0c90d524e90a48f85c97.png) ---
**Author:** ![bramski](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/bramski/32/18511_2.png) [@bramski](https://discuss.circleci.com/u/bramski)\ **Post date:** [June 4, 2018, 12:26am UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/6 "2018-06-04T00:26:07Z")
9 months later. This is such poor design. How often do I have to re-run the failing build? Like 50% of the time. 5 clicks for what should be one click. Boo. ---
**Author:** ![halfer](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/halfer/32/9083_2.png) [@halfer](https://discuss.circleci.com/u/halfer)\ **Post date:** [June 4, 2018, 9:08am UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/7 "2018-06-04T09:08:19Z")
Do you want a hug, @bramski? 🤗 ---
**Author:** ![cjmurphy42](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/cjmurphy42/32/12961_2.png) [@cjmurphy42](https://discuss.circleci.com/u/cjmurphy42)\ **Post date:** [July 10, 2018, 2:45pm UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/8 "2018-07-10T14:45:44Z")
Any update to this? Will this be implemented before CircleCI 1.0 is retired? The current UI for rerunning failed jobs is awkward at best. ---
**Author:** ![benk](https://avatars.discourse-cdn.com/v4/letter/b/ea5d25/32.png) [@benk](https://discuss.circleci.com/u/benk)\ **Post date:** [July 17, 2018, 10:56am UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/9 "2018-07-17T10:56:41Z")
To add another point, for jobs which have not run there is currently no apparent way **at all** to ask them to rebuild. ![Screenshot%20from%202018-07-17%2011-54-04](https://global.discourse-cdn.com/circleci/original/3X/9/2/92c11039b377e11b5563c6b65360762c7954bddb.png) ---
**Author:** ![acidjazz](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/acidjazz/32/16438_2.png) [@acidjazz](https://discuss.circleci.com/u/acidjazz)\ **Post date:** [August 2, 2018, 8:45pm UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/10 "2018-08-02T20:45:30Z")
+1 for this feature, would be super stoked if we could re-run builds ---
**Author:** ![mzgajner](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/mzgajner/32/18799_2.png) [@mzgajner](https://discuss.circleci.com/u/mzgajner)\ **Post date:** [August 8, 2018, 9:59am UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/11 "2018-08-08T09:59:21Z")
This is a pain point for our company as well. Apart from the awkward additional clicking around, it’s problematic because you cannot rerun failed before all the tasks are done. So if task A in a workflow fails and task B is still running, you have to wait for task B to fail/succeed before you can rerun failed tasks. This can severely slow people down. ---
**Author:** ![halfer](https://sea2.discourse-cdn.com/circleci/user_avatar/discuss.circleci.com/halfer/32/9083_2.png) [@halfer](https://discuss.circleci.com/u/halfer)\ **Post date:** [August 8, 2018, 2:54pm UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/12 "2018-08-08T14:54:49Z")
I’ve not tried it, but has anyone looked at making rebuild requests using the API? I imagine `POST: /project/:vcs-type/:username/:project/:build_num/retry` might be a good place to start ([API reference](https://circleci.com/docs/api/v1-reference/)). I appreciate that folks on this thread would prefer a solution integrated into the CircleCI UI, but if it gets people productive again, a quick console solution might suffice. ---
**Author:** ![system](https://global.discourse-cdn.com/circleci/original/3X/5/1/511608acda8e85906e15909b57bd0a46c1dbe291.png) [@system](https://discuss.circleci.com/u/system)\ **Post date:** [November 6, 2018, 2:54pm UTC](https://discuss.circleci.com/t/i-can-only-rebuild-with-ssh/17932/13 "2018-11-06T14:54:51Z")
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.