I booted the "stock" VM with slub_nomerge and after two hours of uptime (and constant Go compilation and testing), the top counts in slabtop for active objects and memory use are:
OBJS ACTIVE USE OBJ SIZE SLABS OBJ/SLAB CACHE SIZE NAME 536896 536869 99% 2.00K 33556 16 1073792K kmalloc-2k 536350 536350 100% 0.02K 3155 170 12620K audit_buffer 536320 536320 100% 0.25K 33520 16 134080K skbuff_head_cache 489099 489099 100% 0.10K 12541 39 50164K buffer_head 78057 75309 96% 0.19K 3717 21 14868K dentry 65110 59084 90% 0.02K 383 170 1532K lsm_inode_cache OBJS ACTIVE USE OBJ SIZE SLABS OBJ/SLAB CACHE SIZE NAME 537104 537097 99% 2.00K 33569 16 1074208K kmalloc-2k 536528 536528 100% 0.25K 33533 16 134132K skbuff_head_cache 465348 397547 85% 0.10K 11932 39 47728K buffer_head 36936 36471 98% 1.15K 1368 27 43776K ext4_inode_cache 76839 56841 73% 0.19K 3659 21 14636K dentry 20988 19716 93% 0.62K 1749 12 13992K inode_cache 536520 536520 100% 0.02K 3156 170 12624K audit_buffer It seems suggestive that the top three by count have almost the same count (and it's a large one). -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1987430 Title: Ubuntu 22.04 kernel 5.15.0-46-generic leaks kernel memory in kmalloc-2k slabs Status in linux package in Ubuntu: Confirmed Bug description: Since updating to kernel 5.15.0-46-generic (package version 5.15.0-46.49), all of our Ubuntu 22.04 LTS servers are leaking kernel memory; our first server with 8 GB of RAM just fatally OOMed, causing us to detect this. Inspection of OOM reports, /proc/meminfo, and /proc/slabinfo says that it's mostly going to unreclaimable kmalloc-2k slabs: Aug 23 12:51:11 cluster kernel: [361299.864757] Unreclaimable slab info: Aug 23 12:51:11 cluster kernel: [361299.864757] Name Used Total [...] Aug 23 12:51:11 cluster kernel: [361299.864924] kmalloc-2k 6676584KB 6676596KB Most of our machines appear to be leaking slab memory at a rate of around 20 to 40 Mbytes/hour, with some machines leaking much faster; the champions are leaking kernel memory at 145 Mbytes/hour and 237 Mbytes/hour. We aren't running any proprietary kernel modules and our only unusual kernel configuration is that we've disabled AppArmor with 'apparmor=0' on the kernel command line. /proc/version_signature: Ubuntu 5.15.0-46.49-generic 5.15.39 Full kernel command line from the Dell R240 system that fatally OOMd: BOOT_IMAGE=/boot/vmlinuz-5.15.0-46-generic root=UUID=3165564f-a2dd-4b39-935b-114f3e23ff54 ro console=ttyS0,115200 console=tty0 apparmor=0 To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1987430/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
