Hi, I seem to be hitting this issue on RHEL 6.5. Is that possible? I've attached strace and the pstack output that I'm getting. Given below are my OS details:
[root@localhost ~]# cat /etc/redhat-release Red Hat Enterprise Linux Server release 6.5 (Santiago) [root@localhost ~]# uname -r 2.6.32-431.el6.x86_64 On Thursday, May 14, 2015 at 6:05:49 PM UTC+5:30, Gil Tene wrote: > > Thanks for identifying the date and change for the fix in RHEL 6 (April > 21, 2015). Strange that the errata makes no mention of it, or it's impact > (it's a security advisory). > > So far, here is what I know when it comes to distro release numbers for > RHEL and it's main cousins (At least in that world, most admins view things > in terms of distro version rather than kernel version. you can look up > associated kernel versions). I'm using good/BAD to mean ("does not have > the bug" / "HAS THE MISSING BARRIER BUG"): > > RHEL 6 (and CentOS 6, and SL 6): 6.0-6.5 are *good*. 6.6 is *BAD*. 6.6.z > is *good*. > RHEL 7 (and CentOS 7, and SL 7): 7.1 is *BAD*. *As of yesterday. there > does not yet appear to be a 7.x fix. * > RHEL 5 (and CentOS 5, and SL 5): All versions are *good* (including 5.11). > > Sadly, 6.6 seems to be a very (the most?) popular version we run into. And > very few have moved to 6.6.z. > > I/we are looking for more info to identify the specific versions affected > in other distros (Ubuntu 12.04 LTS and 14.04 LTS, SLES 12 and 11, Amazon > Linux, Oracle Linux, etc.). So far we've verified that SLES12 kernel > versions 3.12.32-33.1 & above have the fix (but not what versions have the > bug), and that Amazon Linux kernel versions 3.14.35-28.38 & above have the > fix (but not which versions have the bug). > > I will post here as I have more, if you find more info useful to > identifying releases that have the bug and ones that have fixed it, please > do the same. > > On Thursday, May 14, 2015 at 5:24:33 AM UTC-5, Marcin Sobieszczanski wrote: >> >> > More importantly this breakage seems to have >> > been back ported into major distros (e.g. into RHEL 6.6 and its >> cousins, >> > released in October 2014), and the fix for it has only recently been >> back >> > ported (e.g. RHEL 6.6.z and cousins have the fix). >> >> According to the ChangeLogs attached to rpms, it looks like the kernel >> series used in RHEL 6.6 (kernel-2.6.32-504) were affected from the >> start of 6.6 release. It has been fixed only recently in >> kernel-2.6.32-504.16.2.el6 update (21 April, >> https://rhn.redhat.com/errata/RHSA-2015-0864.html) >> >> rpm -qp --changelog kernel-2.6.32-504.16.2.el6.x86_64.rpm | grep >> 'Ensure get_futex_key_refs() always implies a barrier' >> > -- You received this message because you are subscribed to the Google Groups "mechanical-sympathy" group. To unsubscribe from this group and stop receiving emails from it, send an email to mechanical-sympathy+unsubscr...@googlegroups.com. To view this discussion on the web, visit https://groups.google.com/d/msgid/mechanical-sympathy/054061a4-85c4-48ed-85d9-fa12ae2a2d36%40googlegroups.com.
[root@localhost ~]# strace -p 27409 Process 27409 attached - interrupt to quit futex(0xa8c940, FUTEX_WAIT_PRIVATE, 2, NULL