Hi Vivian and all The log attached in Launchpad is not collected in time and valid info was lost. Hope reporter can provide the log before auto-recovered. I commented in: https://bugs.launchpad.net/starlingx/+bug/1831635 Although the issue happens intermittently, but it can always be reproduced once it happens. So it would be easier to root cause it if we can login to the live system. I suppose in Starlingx all the cinder related log are located in: /var/log/containers/ Something like: cinder-api-...log cinder-scheduler-...log cinder-volume-...log Please anybody correct me if I misunderstand this. Thanks. Regards Liang From: Zhu, Vivian Sent: Monday, July 15, 2019 9:55 AM To: Fang, Liang A <liang.a.fang@intel.com>; starlingx-discuss@lists.starlingx.io Subject: RE: Is it possible to run tests inside Intel Is any log helpful to narrow down the issue rather than reproducing it since it is intermittently? Thanks! - Vivian SSG OTC NST Storage Tel: (8621)61167437 From: Fang, Liang A <liang.a.fang@intel.com<mailto:liang.a.fang@intel.com>> Sent: Saturday, July 13, 2019 12:06 AM To: starlingx-discuss@lists.starlingx.io<mailto:starlingx-discuss@lists.starlingx.io> Subject: [Starlingx-discuss] Is it possible to run tests inside Intel Hi I'm trying to fix two bugs that reported by WR folks, but it's hard to reproduce in Intel side manually because it happens intermittently. I'm wondering if the same test scripts can be run regularly in some server inside Intel. So if the issue happens at some time, we can root cause it in the live system. Regards Liang