On 29.03.2018 12:03, Christian Borntraeger wrote: > > > On 03/29/2018 11:37 AM, Thomas Huth wrote: >> The current timeout is set to only three seconds - and considering that >> vring_wait_reply() or rather get_second() is not doing any rounding, >> the real timeout is likely rather 2 seconds in most cases. When the >> host is really badly loaded and we run the guest in TCG mode instead >> of KVM, it's possible that we hit this timeout by mistake. So let's >> increase the timeout to 30 seconds instead to ease this situation (30 >> seconds is also the timeout that is used by the Linux SCSI subsystem >> for example, so this seems to be a sane value for block IO timeout). > > I have never seen this, but cant this also with KVM (e.g. host paging > and the swap disk is busy for some seconds).
Sure, I even assume that Lukáš might have hit this with KVM, and not with TCG. It's just a little bit more unlikely there. > Wouldnt it also qualify for 2.12? Yes, I was just not sure whether we want to rebuild the s390-ccw.img just because of this ... I'd like to leave that decision to you and Cornelia. Thomas
