View Issue Details

IDProjectCategoryView StatusLast Update
0001057Rocky-Linux-8rng-toolspublic2022-11-25 14:55
ReporterRobert Sjöblom Assigned ToLouis Abel  
PrioritynormalSeverityminorReproducibilityhave not tried
Status needinfoResolutionopen 
Summary0001057: rngd.service sometimes fails on upgrade from Rocky 8.6 to 8.7
DescriptionOn dnf update from 8.6 to 8.7, there appears to be an SELinux dependency problem for rngd. This is causing the service to fail since it appears to be updated before the SELinux package. From syslog:

messages:Nov 23 06:10:28 hostname dracut[58625]: *** Including module: rngd ***
messages:Nov 23 06:10:43 hostname rngd[1084]: [rdrand]: Shutting down
messages:Nov 23 06:10:43 hostname rngd[1084]: [jitter]: Shutting down
messages:Nov 23 06:10:43 hostname systemd[1]: rngd.service: Succeeded.
messages:Nov 23 06:10:43 hostname rngd[63797]: Disabling 7: PKCS11 Entropy generator (pkcs11)
messages:Nov 23 06:10:43 hostname rngd[63797]: Disabling 5: NIST Network Entropy Beacon (nist)
messages:Nov 23 06:10:43 hostname rngd[63797]: User 'daemon' not found
messages:Nov 23 06:10:43 hostname systemd[1]: rngd.service: Main process exited, code=exited, status=1/FAILURE
messages:Nov 23 06:10:43 hostname systemd[1]: rngd.service: Failed with result 'exit-code'.
messages:Nov 23 06:10:46 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from search access on the directory /var/lib/sss/mc/passwd. For complete SELinux messages run: sealert -l 3ca404aa-47da-4f94-959b-35d41eceaf96
messages:Nov 23 06:10:46 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from search access on the directory /var/lib/sss/mc/passwd.#012#012***** Plugin restorecon (99.5 confidence) suggests ************************#012#012If you want to fix the label. #012/var/lib/sss/mc/passwd default label should be sssd_public_t.#012Then you can run restorecon. The access attempt may have been stopped due to insufficient permissions to access a parent directory in which case try to change the following command accordingly.#012Do#012# /sbin/restorecon -v /var/lib/sss/mc/passwd#012#012***** Plugin catchall (1.49 confidence) suggests **************************#012#012If you believe that rngd should be allowed search access on the passwd directory by default.#012Then you should report this as a bug.#012You can generate a local policy module to allow this access.#012Do#012allow this access for now by executing:#012# ausearch -c 'rngd' --raw | audit2allow -M my-rngd#012# semodule -X 300 -i my-rngd.pp#012
messages:Nov 23 06:10:46 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from search access on the directory /var/lib/sss/mc/passwd. For complete SELinux messages run: sealert -l 3ca404aa-47da-4f94-959b-35d41eceaf96
messages:Nov 23 06:10:46 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from search access on the directory /var/lib/sss/mc/passwd.#012#012***** Plugin restorecon (99.5 confidence) suggests ************************#012#012If you want to fix the label. #012/var/lib/sss/mc/passwd default label should be sssd_public_t.#012Then you can run restorecon. The access attempt may have been stopped due to insufficient permissions to access a parent directory in which case try to change the following command accordingly.#012Do#012# /sbin/restorecon -v /var/lib/sss/mc/passwd#012#012***** Plugin catchall (1.49 confidence) suggests **************************#012#012If you believe that rngd should be allowed search access on the passwd directory by default.#012Then you should report this as a bug.#012You can generate a local policy module to allow this access.#012Do#012allow this access for now by executing:#012# ausearch -c 'rngd' --raw | audit2allow -M my-rngd#012# semodule -X 300 -i my-rngd.pp#012
messages:Nov 23 06:10:47 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from search access on the directory /var/lib/sss/pipes/nss. For complete SELinux messages run: sealert -l 3ca404aa-47da-4f94-959b-35d41eceaf96
messages:Nov 23 06:10:47 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from search access on the directory /var/lib/sss/pipes/nss.#012#012***** Plugin restorecon (99.5 confidence) suggests ************************#012#012If you want to fix the label. #012/var/lib/sss/pipes/nss default label should be sssd_public_t.#012Then you can run restorecon. The access attempt may have been stopped due to insufficient permissions to access a parent directory in which case try to change the following command accordingly.#012Do#012# /sbin/restorecon -v /var/lib/sss/pipes/nss#012#012***** Plugin catchall (1.49 confidence) suggests **************************#012#012If you believe that rngd should be allowed search access on the nss directory by default.#012Then you should report this as a bug.#012You can generate a local policy module to allow this access.#012Do#012allow this access for now by executing:#012# ausearch -c 'rngd' --raw | audit2allow -M my-rngd#012# semodule -X 300 -i my-rngd.pp#012
messages:Nov 23 06:10:47 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from read access on the file /etc/passwd. For complete SELinux messages run: sealert -l 2ab94b3a-95dc-466c-a32d-5162e4b6a8f3
messages:Nov 23 06:10:47 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from read access on the file /etc/passwd.#012#012***** Plugin catchall (100. confidence) suggests **************************#012#012If you believe that rngd should be allowed read access on the passwd file by default.#012Then you should report this as a bug.#012You can generate a local policy module to allow this access.#012Do#012allow this access for now by executing:#012# ausearch -c 'rngd' --raw | audit2allow -M my-rngd#012# semodule -X 300 -i my-rngd.pp#012
messages:Nov 23 06:10:47 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from search access on the directory /run/dbus/system_bus_socket. For complete SELinux messages run: sealert -l d16511fa-16aa-4e70-a0a9-8125f4b8d336
messages:Nov 23 06:10:47 hostname setroubleshoot[63801]: SELinux is preventing /usr/sbin/rngd from search access on the directory /run/dbus/system_bus_socket.#012#012***** Plugin catchall (100. confidence) suggests **************************#012#012If you believe that rngd should be allowed search access on the system_bus_socket directory by default.#012Then you should report this as a bug.#012You can generate a local policy module to allow this access.#012Do#012allow this access for now by executing:#012# ausearch -c 'rngd' --raw | audit2allow -M my-rngd#012# semodule -X 300 -i my-rngd.pp#012

From dnf.rpm.log:
2022-11-23T06:09:56+0100 SUBDEBUG Upgrade: rng-tools-6.15-1.el8.x86_64
2022-11-23T06:10:43+0100 SUBDEBUG Upgraded: rng-tools-6.14-6.git.b2b7934e.el8_6.x86_64
...
2022-11-23T06:12:47+0100 SUBDEBUG Upgraded: ipa-selinux-4.9.8-8.module+el8.6.0+1050+4989852e.noarch
2022-11-23T06:12:47+0100 SUBDEBUG Upgraded: rpm-plugin-selinux-4.14.3-24.el8_6.x86_64
2022-11-23T06:12:47+0100 SUBDEBUG Upgraded: selinux-policy-targeted-3.14.3-95.el8_6.4.noarch
2022-11-23T06:12:48+0100 SUBDEBUG Upgraded: selinux-policy-3.14.3-95.el8_6.4.noarch
2022-11-23T06:12:50+0100 SUBDEBUG Upgraded: python3-libselinux-2.9-5.el8.x86_64
2022-11-23T06:12:50+0100 SUBDEBUG Upgraded: libselinux-utils-2.9-5.el8.x86_64
2022-11-23T06:13:01+0100 SUBDEBUG Upgraded: libselinux-2.9-5.el8.x86_64

Here we can see rng-tools are upgraded before SELinux package.
TagsNo tags attached.

Activities

Robert Sjöblom

Robert Sjöblom

2022-11-23 09:30

reporter   ~0001222

Restarting rngd after package SELinux has been upgraded succeeds
Robert Sjöblom

Robert Sjöblom

2022-11-24 08:52

reporter   ~0001255

In some cases, restarting rngd fails with "can't find user daemon"; rngd is prevented from reading the file by selinux. Solution is to reinstall selinux-policy package, then restart rngd.
Louis Abel

Louis Abel

2022-11-24 09:39

administrator   ~0001256

I am unable to replicate this issue. After patching and rebooting an 8.6 system, rngd starts up as expected and there are no selinux errors. Reinstalling the selinux-policy package leads me to believe there's either a configuration issue or there is possibly an edge case you've ran into. Can you provide your entire dnf.rpm.log of the day you ran dnf update?

[root@router scsi]# uname -r
4.18.0-372.26.1.el8_6.x86_64
[root@router scsi]# dnf update -y -q
[root@router scsi]# rpm -q rng-tools
rng-tools-6.15-1.el8.x86_64
[root@router scsi]# rpm -q selinux-policy
selinux-policy-3.14.3-108.el8.noarch
[root@router scsi]# systemctl status rngd
● rngd.service - Hardware RNG Entropy Gatherer Daemon
   Loaded: loaded (/usr/lib/systemd/system/rngd.service; enabled; vendor preset: enabled)
   Active: active (running) since Tue 2022-11-15 11:06:42 MST; 1 weeks 1 days ago
 Main PID: 333967 (rngd)
    Tasks: 5 (limit: 409620)
   Memory: 1.9M
   CGroup: /system.slice/rngd.service
           └─333967 /usr/sbin/rngd -f --fill-watermark=0 -x pkcs11 -x nist -D daemon:daemon

Nov 15 11:06:42 router systemd[1]: Started Hardware RNG Entropy Gatherer Daemon.
Nov 15 11:06:42 router rngd[333967]: Disabling 7: PKCS11 Entropy generator (pkcs11)
Nov 15 11:06:42 router rngd[333967]: Disabling 5: NIST Network Entropy Beacon (nist)
Nov 15 11:06:42 router rngd[333967]: Initializing available sources
Nov 15 11:06:42 router rngd[333967]: [hwrng ]: Initialized
Nov 15 11:06:42 router rngd[333967]: [rdrand]: Initialization Failed
Nov 15 11:06:42 router rngd[333967]: [jitter]: Initializing AES buffer
Nov 15 11:06:47 router rngd[333967]: [jitter]: Enabling JITTER rng support
Nov 15 11:06:47 router rngd[333967]: [jitter]: Initialized
Nov 15 11:06:47 router rngd[333967]: Process privileges have been dropped to 2:2
[root@router scsi]# init 6
[root@router ~]# systemctl status rngd
● rngd.service - Hardware RNG Entropy Gatherer Daemon
   Loaded: loaded (/usr/lib/systemd/system/rngd.service; enabled; vendor preset: enabled)
   Active: active (running) since Thu 2022-11-24 02:15:32 MST; 1min 15s ago
 Main PID: 1537 (rngd)
    Tasks: 5 (limit: 409712)
   Memory: 6.5M
   CGroup: /system.slice/rngd.service
           └─1537 /usr/sbin/rngd -f --fill-watermark=0 -x pkcs11 -x nist -D daemon:daemon

Nov 24 02:15:32 router systemd[1]: Started Hardware RNG Entropy Gatherer Daemon.
Nov 24 02:15:32 router rngd[1537]: Disabling 7: PKCS11 Entropy generator (pkcs11)
Nov 24 02:15:32 router rngd[1537]: Disabling 5: NIST Network Entropy Beacon (nist)
Nov 24 02:15:32 router rngd[1537]: Initializing available sources
Nov 24 02:15:32 router rngd[1537]: [hwrng ]: Initialized
Nov 24 02:15:32 router rngd[1537]: [rdrand]: Initialization Failed
Nov 24 02:15:32 router rngd[1537]: [jitter]: Initializing AES buffer
Nov 24 02:15:37 router rngd[1537]: [jitter]: Enabling JITTER rng support
Nov 24 02:15:37 router rngd[1537]: [jitter]: Initialized
Nov 24 02:15:37 router rngd[1537]: Process privileges have been dropped to 2:2
[root@router ~]# audit2why < /var/log/audit/audit.log
[root@router ~]#

[root@router ~]# grep -En 'rng|selinux' /tmp/dnf.rpm.log
5:2022-11-24T01:56:03-0700 SUBDEBUG Upgrade: libselinux-2.9-6.el8.x86_64
44:2022-11-24T01:56:08-0700 SUBDEBUG Upgrade: libselinux-utils-2.9-6.el8.x86_64
95:2022-11-24T01:56:19-0700 SUBDEBUG Upgrade: libselinux-devel-2.9-6.el8.x86_64
144:2022-11-24T01:57:05-0700 SUBDEBUG Upgrade: python3-libselinux-2.9-6.el8.x86_64
196:2022-11-24T01:57:21-0700 SUBDEBUG Upgrade: rpm-plugin-selinux-4.14.3-24.el8_7.x86_64
197:2022-11-24T01:57:21-0700 SUBDEBUG Upgrade: selinux-policy-3.14.3-108.el8.noarch
198:2022-11-24T01:57:39-0700 SUBDEBUG Upgrade: selinux-policy-targeted-3.14.3-108.el8.noarch
247:2022-11-24T01:58:38-0700 SUBDEBUG Upgrade: ipa-selinux-4.9.10-3.module+el8.7.0+1074+aae18f3a.noarch
398:2022-11-24T02:05:30-0700 SUBDEBUG Upgrade: rng-tools-6.15-1.el8.x86_64
522:2022-11-24T02:06:20-0700 SUBDEBUG Upgraded: libselinux-devel-2.9-5.el8.x86_64
540:2022-11-24T02:06:22-0700 SUBDEBUG Upgraded: ipa-selinux-4.9.8-6.module+el8.6.0+797+07647629.noarch
582:2022-11-24T02:06:42-0700 SUBDEBUG Upgraded: rng-tools-6.14-6.git.b2b7934e.el8_6.x86_64
814:2022-11-24T02:08:36-0700 SUBDEBUG Upgraded: rpm-plugin-selinux-4.14.3-24.el8_6.x86_64
815:2022-11-24T02:08:36-0700 SUBDEBUG Upgraded: selinux-policy-targeted-3.14.3-95.el8_6.4.noarch
816:2022-11-24T02:08:37-0700 SUBDEBUG Upgraded: selinux-policy-3.14.3-95.el8_6.4.noarch
842:2022-11-24T02:08:39-0700 SUBDEBUG Upgraded: python3-libselinux-2.9-5.el8.x86_64
848:2022-11-24T02:08:39-0700 SUBDEBUG Upgraded: libselinux-utils-2.9-5.el8.x86_64
910:2022-11-24T02:08:56-0700 SUBDEBUG Upgraded: libselinux-2.9-5.el8.x86_64
Robert Sjöblom

Robert Sjöblom

2022-11-25 14:55

reporter   ~0001288

We've only seen this issue on a few of the servers in the fleet, namely the postgres servers. We're currently running ~900 postgres servers, out of ~1600 machines total. Around 10 of them have been affected by this issue, so it seems likely that it's an edge condition of some kind. We use ansible to configure all our servers, and so they should be exactly similar in configuration.

The servers we have seen this issue on were initially installed with CentOS 8, and later converted to Rocky using Rocky's conversion script. It's possible that it's related, but it's also a fact that the large majority of servers are in this state. We have much fewer new servers with a clean Rocky 8 install.


When looking into the dnf rpm log during the upgrade window, we found an SELinux policy error due to a memory allocation failure.
2022-11-24T05:43:00+0100 SUBDEBUG Upgrade: selinux-policy-targeted-3.14.3-108.el8.noarch
2022-11-24T05:43:38+0100 INFO SELinux: Could not load policy file /etc/selinux/targeted/policy/policy.31: Cannot allocate memory
2022-11-24T05:43:38+0100 SUBDEBUG Upgrade: sssd-dbus-2.7.3-4.el8_7.1.x86_64


The system journal at the same time shows the following:
Nov 24 05:43:37 hostname kernel: load_policy: page allocation failure: order:4, mode:0x60c0c0(GFP_KERNEL|__GFP_COMP|__GFP_ZERO), nodemask=(null),cpuset=/,mems_allowed=0
Nov 24 05:43:37 hostname kernel: CPU: 1 PID: 2351056 Comm: load_policy Kdump: loaded Not tainted 4.18.0-372.9.1.el8.x86_64 #1
Nov 24 05:43:37 hostname kernel: Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 11/12/2020
Nov 24 05:43:37 hostname kernel: Call Trace:
Nov 24 05:43:37 hostname kernel: dump_stack+0x41/0x60
Nov 24 05:43:37 hostname kernel: warn_alloc.cold.119+0x7b/0x111
Nov 24 05:43:37 hostname kernel: ? _cond_resched+0x15/0x30
Nov 24 05:43:37 hostname kernel: ? __alloc_pages_direct_compact+0x157/0x160
Nov 24 05:43:37 hostname kernel: __alloc_pages_slowpath+0xc7e/0xcc0
Nov 24 05:43:37 hostname kernel: ? type_read+0x160/0x160
Nov 24 05:43:37 hostname kernel: __alloc_pages_nodemask+0x2db/0x310
Nov 24 05:43:37 hostname kernel: kmalloc_order+0x28/0x90
Nov 24 05:43:37 hostname kernel: kmalloc_order_trace+0x1d/0xa0
Nov 24 05:43:37 hostname kernel: ? type_read+0x160/0x160
Nov 24 05:43:37 hostname kernel: __kmalloc+0x1ff/0x250
Nov 24 05:43:37 hostname kernel: ? type_read+0x160/0x160
Nov 24 05:43:37 hostname kernel: hashtab_init+0x5d/0x80
Nov 24 05:43:37 hostname kernel: policydb_read+0x2e3/0x1230
Nov 24 05:43:37 hostname kernel: security_load_policy+0xa8/0x5e0
Nov 24 05:43:37 hostname kernel: ? copy_user_generic_unrolled+0x32/0xc0
Nov 24 05:43:37 hostname kernel: sel_write_load+0xde/0x1a0
Nov 24 05:43:37 hostname kernel: vfs_write+0xa5/0x1a0
Nov 24 05:43:37 hostname kernel: ksys_write+0x4f/0xb0
Nov 24 05:43:37 hostname kernel: do_syscall_64+0x5b/0x1a0
Nov 24 05:43:37 hostname kernel: entry_SYSCALL_64_after_hwframe+0x65/0xca
Nov 24 05:43:37 hostname kernel: RIP: 0033:0x7fa9bd1f2bc8
Nov 24 05:43:37 hostname kernel: Code: 89 02 48 c7 c0 ff ff ff ff eb b3 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 8d 05 55 4b 2a 00 8b 00 85 c0 75 17 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 58 c3 0f 1f 80 00 00 00 00 41 54 49 89 d4 55
Nov 24 05:43:37 hostname kernel: RSP: 002b:00007ffe2428f228 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
Nov 24 05:43:37 hostname kernel: RAX: ffffffffffffffda RBX: 00007ffe2428f230 RCX: 00007fa9bd1f2bc8
Nov 24 05:43:37 hostname kernel: RDX: 0000000000833419 RSI: 00007fa9af240000 RDI: 0000000000000004
Nov 24 05:43:37 hostname kernel: RBP: 0000000000000004 R08: 000055cc2d2e22a0 R09: 00007fa9bd252d40
Nov 24 05:43:37 hostname kernel: R10: 0000000000000000 R11: 0000000000000246 R12: 00007fa9af240000
Nov 24 05:43:37 hostname kernel: R13: 0000000000833419 R14: 000000000000000f R15: 0000000000000003
Nov 24 05:43:37 hostname kernel: Mem-Info:
Nov 24 05:43:37 hostname kernel: active_anon:17640 inactive_anon:182469 isolated_anon:0
                                                           active_file:812856 inactive_file:1292224 isolated_file:0
                                                           unevictable:0 dirty:6 writeback:0
                                                           slab_reclaimable:170338 slab_unreclaimable:66719
                                                           mapped:57643 shmem:41017 pagetables:4920 bounce:0
                                                           free:135449 free_pcp:0 free_cma:0
Nov 24 05:43:37 hostname kernel: Node 0 active_anon:70560kB inactive_anon:729876kB active_file:3251424kB inactive_file:5168896kB unevictable:0kB isolated(anon):0kB isolated(file):0kB mapped:230572kB dirty:24kB writeback:0kB shmem:164068kB shmem_thp: 0kB shmem_pmdmapped: 0kB anon_thp: 12288kB writeback_tmp:0kB kernel_stack:8400kB pagetables:19680kB all_unreclaimable? no
Nov 24 05:43:37 hostname kernel: Node 0 DMA free:13312kB min:64kB low:80kB high:96kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB writepending:0kB present:15988kB managed:15360kB mlocked:0kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB
Nov 24 05:43:37 hostname kernel: lowmem_reserve[]: 0 2768 15745 15745 15745
Nov 24 05:43:37 hostname kernel: Node 0 DMA32 free:434312kB min:11972kB low:14964kB high:17956kB active_anon:7344kB inactive_anon:347200kB active_file:655228kB inactive_file:936252kB unevictable:0kB writepending:16kB present:3129152kB managed:2867008kB mlocked:0kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB
Nov 24 05:43:37 hostname kernel: lowmem_reserve[]: 0 0 12977 12977 12977
Nov 24 05:43:37 hostname kernel: Node 0 Normal free:94172kB min:55544kB low:69428kB high:83312kB active_anon:63216kB inactive_anon:382852kB active_file:2596196kB inactive_file:4232836kB unevictable:0kB writepending:8kB present:13631488kB managed:13297412kB mlocked:0kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB
Nov 24 05:43:37 hostname kernel: lowmem_reserve[]: 0 0 0 0 0
Nov 24 05:43:37 hostname kernel: Node 0 DMA: 0*4kB 0*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 1*1024kB (U) 2*2048kB (UM) 2*4096kB (M) = 13312kB
Nov 24 05:43:37 hostname kernel: Node 0 DMA32: 24038*4kB (UME) 22093*8kB (UME) 9551*16kB (UME) 293*32kB (UME) 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 435088kB
Nov 24 05:43:37 hostname kernel: Node 0 Normal: 8267*4kB (UMEH) 1192*8kB (UMEH) 2403*16kB (UMEH) 391*32kB (UH) 10*64kB (H) 3*128kB (H) 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 94588kB
Nov 24 05:43:37 hostname kernel: Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=1048576kB
Nov 24 05:43:37 hostname kernel: Node 0 hugepages_total=2400 hugepages_free=287 hugepages_surp=0 hugepages_size=2048kB
Nov 24 05:43:37 hostname kernel: 2140592 total pagecache pages
Nov 24 05:43:37 hostname kernel: 65 pages in swap cache
Nov 24 05:43:37 hostname kernel: Swap cache stats: add 217664, delete 217599, find 49323/56024
Nov 24 05:43:37 hostname kernel: Free swap = 1385212kB
Nov 24 05:43:37 hostname kernel: Total swap = 2064380kB
Nov 24 05:43:37 hostname kernel: 4194157 pages RAM
Nov 24 05:43:37 hostname kernel: 0 pages HighMem/MovableOnly
Nov 24 05:43:37 hostname kernel: 149212 pages reserved
Nov 24 05:43:37 hostname kernel: 0 pages hwpoisoned
Nov 24 05:43:37 hostname kernel: SELinux: failed to load policy
Nov 24 05:43:38 hostname dnf-automatic[2350061]: SELinux: Could not load policy file /etc/selinux/targeted/policy/policy.31: Cannot allocate memory
Nov 24 05:43:38 hostname dnf-automatic[2350061]: load_policy: Can't load policy: Cannot allocate memory
Nov 24 05:43:38 hostname dbus-daemon[1100]: [system] Reloaded configuration
Nov 24 05:43:38 hostname dbus-daemon[1100]: [system] Reloaded configuration
Nov 24 05:43:38 hostname dbus-daemon[1100]: [system] Reloaded configuration
Nov 24 05:43:38 hostname dbus-daemon[1100]: [system] Reloaded configuration
Nov 24 05:43:40 hostname dbus-daemon[1100]: [system] Reloaded configuration
Nov 24 05:43:40 hostname dbus-daemon[1100]: [system] Reloaded configuration
Nov 24 05:43:41 hostname systemd-udevd[2351902]: Using default interface naming scheme 'rhel-8.0'.
Nov 24 05:43:42 hostname systemd[1]: Reloading.
Nov 24 05:43:43 hostname systemd[1]: Reloading.
Nov 24 05:43:47 hostname NetworkManager[1238]: <info> [1669265027.8825] manager: kernel firmware directory '/lib/firmware' changed


Due to this server being a database host, we have allocated hugepages and disabled overcommit. We have 2400 * 2048 kB memory allocated for hugepages and 11076044 kB available for userspace applications (CommitLimit). Total memory on the machine is, according to /proc/meminfo, 16179780 kB which should leave 188536 kB for the kernel. Perhaps this is not enough?


Here's our cat /proc/meminfo for one of the affected servers:
MemTotal: 16179780 kB
MemFree: 384992 kB
MemAvailable: 9381080 kB
Buffers: 3704 kB
Cached: 8809540 kB
SwapCached: 276 kB
Active: 3656528 kB
Inactive: 5692040 kB
Active(anon): 83920 kB
Inactive(anon): 611336 kB
Active(file): 3572608 kB
Inactive(file): 5080704 kB
Unevictable: 0 kB
Mlocked: 0 kB
SwapTotal: 2064380 kB
SwapFree: 1387004 kB
Dirty: 92 kB
Writeback: 0 kB
AnonPages: 531136 kB
Mapped: 257248 kB
Shmem: 189552 kB
KReclaimable: 686400 kB
Slab: 953032 kB
SReclaimable: 686400 kB
SUnreclaim: 266632 kB
KernelStack: 8384 kB
PageTables: 19500 kB
NFS_Unstable: 0 kB
Bounce: 0 kB
WritebackTmp: 0 kB
CommitLimit: 11076044 kB
Committed_AS: 2940032 kB
VmallocTotal: 34359738367 kB
VmallocUsed: 0 kB
VmallocChunk: 0 kB
Percpu: 101888 kB
HardwareCorrupted: 0 kB
AnonHugePages: 10240 kB
ShmemHugePages: 0 kB
ShmemPmdMapped: 0 kB
FileHugePages: 0 kB
FilePmdMapped: 0 kB
HugePages_Total: 2400
HugePages_Free: 287
HugePages_Rsvd: 9
HugePages_Surp: 0
Hugepagesize: 2048 kB
Hugetlb: 4915200 kB
DirectMap4k: 1275712 kB
DirectMap2M: 12355584 kB
DirectMap1G: 5242880 kB

Issue History

Date Modified Username Field Change
2022-11-23 09:30 Robert Sjöblom New Issue
2022-11-23 09:30 Robert Sjöblom Note Added: 0001222
2022-11-24 08:52 Robert Sjöblom Note Added: 0001255
2022-11-24 09:39 Louis Abel Assigned To => Louis Abel
2022-11-24 09:39 Louis Abel Status new => needinfo
2022-11-24 09:39 Louis Abel Note Added: 0001256
2022-11-25 14:55 Robert Sjöblom Note Added: 0001288