https://jira.whamcloud.com This file is an XML representation of an issue en-us 9.4.14 940014 05-12-2023 https://jira.whamcloud.com/browse/LU-7781 Lustre <ul> <li>It was found that the Linux kernel's keys subsystem did not correctly<br/> garbage collect uninstantiated keyrings. A local attacker could use this<br/> flaw to crash the system or, potentially, escalate their privileges on the<br/> system. (CVE-2015-7872, Important)</li> </ul> <ul> <li>A flaw was found in the way the Linux kernel handled IRET faults during<br/> the processing of NMIs. An unprivileged, local user could use this flaw to<br/> crash the system or, potentially (although highly unlikely), escalate their<br/> privileges on the system. (CVE-2015-5157, Moderate)</li> </ul> <p>This update also fixes the following bugs:</p> <ul> <li>Previously, processing packets with a lot of different IPv6 source<br/> addresses caused the kernel to return warnings concerning soft-lockups due<br/> to high lock contention and latency increase. With this update, lock<br/> contention is reduced by backing off concurrent waiting threads on the<br/> lock. As a result, the kernel no longer issues warnings in the described<br/> scenario. (BZ#1285370)</li> </ul> <ul> <li>Prior to this update, block device readahead was artificially limited.<br/> As a consequence, the read performance was poor, especially on RAID<br/> devices. Now, per-device readahead limits are used for each device instead<br/> of a global limit. As a result, read performance has improved, especially<br/> on RAID devices. (BZ#1287550)</li> </ul> <ul> <li>After injecting an EEH error, the host was previously not recovering and<br/> observing I/O hangs in HTX tool logs. This update makes sure that when one<br/> or both of EEH_STATE_MMIO_ACTIVE and EEH_STATE_MMIO_ENABLED flags is marked<br/> in the PE state, the PE's IO path is regarded as enabled as well. As a<br/> result, the host no longer hangs and recovers as expected. (BZ#1289101)</li> </ul> <ul> <li>The genwqe device driver was previously using the GFP_ATOMIC flag for<br/> allocating consecutive memory pages from the kernel's atomic memory pool,<br/> even in non-atomic situations. This could lead to allocation failures<br/> during memory pressure. With this update, the genwqe driver's memory<br/> allocations use the GFP_KERNEL flag, and the driver can allocate memory<br/> even during memory pressure situations. (BZ#1289450)</li> </ul> <ul> <li>The nx842 co-processor for IBM Power Systems could in some circumstances<br/> provide invalid data due to a data corruption bug during uncompression.<br/> With this update, all compression and uncompression calls to the nx842<br/> co-processor contain a cyclic redundancy check (CRC) flag, which forces all<br/> compression and uncompression operations to check data integrity and<br/> prevents the co-processor from providing corrupted data. (BZ#1289451)</li> </ul> <ul> <li>A failed "updatepp" operation on the little-endian variant of IBM Power<br/> Systems could previously cause a wrong hash value to be used for the next<br/> hash insert operation in the page table. This could result in a missing<br/> hash pte update or invalidate operation, potentially causing memory<br/> corruption. With this update, the hash value is always recalculated after a<br/> failed "updatepp" operation, avoiding memory corruption. (BZ#1289452)</li> </ul> <ul> <li>Large Receive Offload (LRO) flag disabling was not being propagated<br/> downwards from above devices in vlan and bond hierarchy, breaking the flow<br/> of traffic. This problem has been fixed and LRO flags now propagate<br/> correctly. (BZ#1292072)</li> </ul> <ul> <li>Due to rounding errors in the CPU frequency of the intel_pstate driver,<br/> the CPU frequency never reached the value requested by the user. A kernel<br/> patch has been applied to fix these rounding errors. (BZ#1296276)</li> </ul> <ul> <li>When running several containers (up to 100), reports of hung tasks were<br/> previously reported. This update fixes the AB-BA deadlock in the<br/> dm_destroy() function, and the hung reports no longer occur. (BZ#1296566)</li> </ul> <p>Bugs fixed (<a href="https://bugzilla.redhat.com/):" class="external-link" target="_blank" rel="nofollow noopener">https://bugzilla.redhat.com/):</a></p> <p>1259577 - CVE-2015-5157 kernel: x86-64: IRET faults during NMIs processing<br/> 1272371 - CVE-2015-7872 kernel: Keyrings crash triggerable by unprivileged user</p> LU-7781

kernel update [RHEL7.2 3.10.0-327.10.1.el7]

Bug Minor Resolved Fixed Bob Glossman Bob Glossman Tue, 16 Feb 2016 16:08:24 +0000 Fri, 1 Jul 2016 18:48:39 +0000 Mon, 14 Mar 2016 03:20:31 +0000 Lustre 2.9.0 0 3 <p>Bob Glossman (bob.glossman@intel.com) uploaded a new patch: <a href="http://review.whamcloud.com/18478" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/18478</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-7781" title="kernel update [RHEL7.2 3.10.0-327.10.1.el7]" class="issue-link" data-issue-key="LU-7781"><del>LU-7781</del></a> kernel: kernel update RHEL7.2 <span class="error">&#91;3.10.0-327.10.1.el7&#93;</span><br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: aa6ef756b3609784a1b1bd850f3344c8402bf943</p> <p>Oleg Drokin (oleg.drokin@intel.com) merged in patch <a href="http://review.whamcloud.com/18478/" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/18478/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-7781" title="kernel update [RHEL7.2 3.10.0-327.10.1.el7]" class="issue-link" data-issue-key="LU-7781"><del>LU-7781</del></a> kernel: kernel update RHEL7.2 <span class="error">&#91;3.10.0-327.10.1.el7&#93;</span><br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: bd7facb7922f5a494ba3af04daa221be96592c56</p> Related LU-7581 Development Rank 1|hzy1dr: Rank (Obsolete) 9223372036854775807 Severity