SuperMicro X11SDV-8C-TP8F
我已经在 Intel 上安装了 Ubuntu Server 22.04 内核 5.15 及以上版本D-2146NT
(我也尝试过 ubuntu 22.04 内核 5.17 并安装 ubuntu 20.04 内核 5.15 和 5.14,结果相同。为了避免此问题,我需要使用早于 5.14 的内核)。
服务器启动空闲时 RAM 使用率很高,我无法解释。
在启动时没有任何活动服务,我使用了近 6GiB 的 RAM:
# free -m
total used free shared buff/cache available
Mem: 15663 5998 8936 155 728 9213
Swap: 0 0 0
使用 ps 时 RSS 内存总量不合适:它显示使用的内存少于 1GiB:
# ps aux | awk '{sum+=$6} END {print sum / 1024}'
944.879
这是 meminfo 输出
cat /proc/meminfo
MemTotal: 16039044 kB
MemFree: 8486372 kB
MemAvailable: 9607120 kB
Buffers: 79440 kB
Cached: 1300092 kB
SwapCached: 0 kB
Active: 575856 kB
Inactive: 1141004 kB
Active(anon): 24184 kB
Inactive(anon): 369940 kB
Active(file): 551672 kB
Inactive(file): 771064 kB
Unevictable: 27884 kB
Mlocked: 27884 kB
SwapTotal: 2097148 kB
SwapFree: 2097148 kB
Dirty: 36 kB
Writeback: 0 kB
AnonPages: 365256 kB
Mapped: 304856 kB
Shmem: 50208 kB
KReclaimable: 110308 kB
Slab: 455680 kB
SReclaimable: 110308 kB
SUnreclaim: 345372 kB
KernelStack: 7664 kB
PageTables: 6352 kB
NFS_Unstable: 0 kB
Bounce: 0 kB
WritebackTmp: 0 kB
CommitLimit: 10116668 kB
Committed_AS: 2497484 kB
VmallocTotal: 34359738367 kB
VmallocUsed: 31940 kB
VmallocChunk: 0 kB
Percpu: 12800 kB
HardwareCorrupted: 0 kB
AnonHugePages: 4096 kB
ShmemHugePages: 0 kB
ShmemPmdMapped: 0 kB
FileHugePages: 0 kB
FilePmdMapped: 0 kB
HugePages_Total: 0
HugePages_Free: 0
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 2048 kB
Hugetlb: 0 kB
DirectMap4k: 315700 kB
DirectMap2M: 11919360 kB
DirectMap1G: 6291456 kB
这是slabtop
标题:
Active / Total Objects (% used) : 1095487 / 1172800 (93.4%)
Active / Total Slabs (% used) : 28258 / 28258 (100.0%)
Active / Total Caches (% used) : 121 / 176 (68.8%)
Active / Total Size (% used) : 401911.16K / 415423.19K (96.7%)
Minimum / Average / Maximum Object : 0.01K / 0.35K / 10.62K
这是df -h
输出:
df -h
Filesystem Size Used Avail Use% Mounted on
tmpfs 1.6G 50M 1.5G 4% /run
/dev/mapper/ubuntu--vg-ubuntu--lv 14G 3.8G 9.0G 30% /
tmpfs 7.7G 20K 7.7G 1% /dev/shm
tmpfs 5.0M 0 5.0M 0% /run/lock
/dev/sda2 2.0G 127M 1.7G 7% /boot
/dev/mapper/vg0-var_log 9.8G 127M 9.2G 2% /var/log
/dev/mapper/vg0-tmp 9.8G 68K 9.3G 1% /tmp
/dev/mapper/vg0-var_tmp 9.8G 96K 9.3G 1% /var/tmp
/dev/mapper/vg0-var_lib_docker 1.7T 11G 1.6T 1% /var/lib/docker
tmpfs 1.6G 0 1.6G 0% /run/user/2010
tmpfs 1.6G 0 1.6G 0% /run/user/2023
答案1
内存使用量是由于irdma
内核模块造成的。我还不知道它是否按预期工作,或者是否是一个错误。
我找到了 irdma 文档这里