Skip to main content
Share Your Experience: Take the 2024 Developer Survey
35 votes
Accepted

What could explain unknown I/O on new 16TB HDD on a Debian server?

The background IO load you noticed is due to ext4 deferred inode table allocation. From mke2fs man page: lazy_itable_init[= <0 to disable, 1 to enable>] If enabled and the uninit_bg feature ...
shodanshok's user avatar
  • 50.9k
26 votes

How does vm.overcommit_memory work?

This is an old question with a well-established answer, but I think there's more to add. First of all, when vm.overcommit_memory = 0, the vm.overcommit_ratio value is irrelevant. The kernel will use ...
d4nyll's user avatar
  • 384
22 votes

GRUB "Some modules may be missing from core image" warning

Had the same thing today. Turns out it's caused by grub-probe trying to access partitions through /dev/sda, which is not cache-coherent with /dev/sda1 (and sda2 etcetera). You can fix it using ...
Fruit's user avatar
  • 221
20 votes
Accepted

How to cleanup /usr/lib/modules and /usr/lib/x86_64-linux-gnu?

I needed to do this, as with over a year of using Linux on a Chromebook, /usr/lib/modules was taking over 10GiB of disk space (which is at a premium on this device) and I am only ever going to be ...
Marcus Harrison's user avatar
19 votes

how to tell if NIC has multiqueue enabled?

ethtool -l <interface> will display the status of queues associated with an interface, if that interface's driver supports such a thing. In ethtool-land, multiqueue is indicated by "channels". ...
Jesse Adelman's user avatar
19 votes
Accepted

Why can't QEMU allocate the memory if the Linux caches are too big?

Not all cached data can be discarded immediately. For example, cached dirty pages have to be written back to disk before they can be removed from RAM. You have no swap, so until those writes complete, ...
David Schwartz's user avatar
18 votes
Accepted

Disabling rp_filter on one interface

Info there: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/tree/Documentation/networking/ip-sysctl.txt?h=v4.9#n1090 Note the last sentence that would explain your attempts: ...
A.B's user avatar
  • 12.9k
16 votes

Prevent the console from clearing the screen?

With systemd things are different. See article Stop Clearing My God Damned Console. In short: mkdir /etc/systemd/system/[email protected] cat >/etc/systemd/system/[email protected]/noclear.conf &...
Tino's user avatar
  • 1,163
12 votes

Kernel stack trace to source code lines

Install kernel-debuginfo Download decode_stacktrace.sh which is in the kernel source tree. Make stack dump output useful again. # ./decode_stacktrace.sh /usr/lib/debug/lib/modules/`uname -r`/...
wenjianhn's user avatar
  • 261
10 votes

rm on a directory with millions of files

TLDR: use rsync -a --delete emptyfolder/ x. This question has 50k views, and quite a few answers, but nobody seems to have benchmarked all the different replies. There's one link to an external ...
adamf's user avatar
  • 201
9 votes

kube-proxy won't start in Minikube because of permission denied issue with /proc/sys/net/netfilter/nf_conntrack_max

change maxPerCore to 0 in configMap of kube-proxy to leave the limit as-is and ignore conntrack-min
Mehdi Shakeri's user avatar
8 votes

How does CPU affinity interact with cgroups in Linux?

From the cpusets documentation: Calls to sched_setaffinity are filtered to just those CPUs allowed in that task's cpuset. This implies that CPU affinity masks are intersected with the cpus in ...
Edd Barrett's user avatar
8 votes

Can't run docker-container 'failed to create endpoint frosty_varahamihira on network bridge'

install the linux-modules-extra-raspi package to get the veth module.
mayostard's user avatar
7 votes

TIME_WAIT consumes file descriptors?

The file descriptor is used by application to read/write from the socket. Thus if the application call close(), the file descriptor is immediately released. On the other hand, if the application call ...
James Gan's user avatar
  • 396
7 votes

Best way to disable swap in Linux

On Raspbian 10 (Buster), the clean answer would be: To disable it until the next reboot, as stated in */etc/fstab*: sudo /sbin/dphys-swapfile swapoff To disable swap on boot: sudo systemctl ...
Michaël GENAY's user avatar
6 votes
Accepted

grubby fatal error: unable to find a suitable template

Somewhat of a late response, but just faced the same problem with CentOS 7 minimal install. Solution was to run: grub2-mkconfig -o /boot/grub2/grub.cfg after upgrading the kernel. Note that does not ...
isedev's user avatar
  • 249
6 votes
Accepted

Why am I having issues upgrading the kernel on a fresh Ubuntu 16.04 installation?

You can't upgrade the kernel because this isn't a real virtual machine. Instead, it's a container using OpenVZ containers, which are rather old, and still use a kernel ABI/version from 2010. You will ...
Michael Hampton's user avatar
6 votes
Accepted

Initial SYN timeout on linux

The initial retransmission timeout setting is hardcoded in the kernel to be 1 second in modern versions: https://elixir.bootlin.com/linux/v5.9.11/source/include/net/tcp.h#L142 #define TCP_TIMEOUT_INIT ...
Juraj Martinka's user avatar
5 votes

Why having more and faster cores makes my multithreaded software slower?

Sure looks like a NUMA effect when multiple sockets degrades performance drastically. perf is very useful. Already in the perf report, you can see native_queued_spin_lock_slowpath taking 35%, which ...
John Mahowald's user avatar
5 votes
Accepted

Debian: Kernel update using APT - reboot required to take effect?

Yes, usually the kernel update is not done in-place so a reboot is needed to update the running kernel. You might want to try needrestart via sudo needrestart -kr l It will check if the correct ...
boppy's user avatar
  • 521
5 votes
Accepted

What are the options to monitoring page cache usage?

In /proc/meminfo you have a number of memory parameters for the system. In particular, the relevant values should be Cached: (Memory used by the page cache) and Dirty: (not just those of the page ...
Ángel's user avatar
  • 912
4 votes

Kernel stack trace to source code lines

With gdb, you can also use this command to find the line number quickly: (gdb) list *(some_function+0x12c)
accessory's user avatar
  • 151
4 votes
Accepted

Linode Kernel Headers

This is old and has long since been solved but I just logged in for the first time in a long time and noticed it has no answer so I am providing one to add closure. In order to install the r1soft ...
digitaladdictions's user avatar
4 votes

Find out which task is generating a lot of context switches on linux

In newer kernel version sudo perf record -e context-switches -a # record the events # then ctrl+c sudo perf report # inspect the result This will give you the exactly result about context-...
snyh's user avatar
  • 141
4 votes

Why are connections in FIN_WAIT2 state not closed by the Linux kernel?

If the socket is shutdown(), but not close() yet, the socket will stay in FIN_WAIT2 state. And since the application still owns the file descriptor, the kernel wouldn't bother to clean up.
L. Yan's user avatar
  • 41
4 votes
Accepted

Rasbian(ARM) in Qemu emulator with Kernel panic - not syncing: Attempted to kill init

Alright. I have to answer this question before anyone else. I followed this tutorial. https://blog.3mdeb.com/2015/2015-12-30-emulate-rapberry-pi-2-in-qemu/ The key is that Changing ld.so.preload Then ...
dotslash's user avatar
  • 239
4 votes

Can I delete folders in /lib/modules/ after calling "apt-get purge"

Ok so I decided to just take the risk and manually delete the folder under /lib/modules and had no problem rebooting my server.
Dalton Tan's user avatar
4 votes

Can I delete folders in /lib/modules/ after calling "apt-get purge"

You can safely remove the corresponding version of packages to the version of the kernel that you removed. In your example because you are missing the corresponding kernel versions for the following, ...
NotoriousPyro's user avatar
4 votes
Accepted

How are Unikernels different from Containers?

Containers share kernel space with the other containers on a physical hypervisor. Generally, unikernels do not. All the functional differences are basically manifestations of that root difference. ...
Omar Buhidma's user avatar
4 votes

Is there anyway to prevent malicious virtual machines from crashing host machine?

Let me formulate it in another way: if a malicious VM crashes the host, the VM infrastructure is garbage and has a security hole. It is not possible, it is a CORE function of a VM infrastructure to ...
TomTom's user avatar
  • 52k

Only top scored, non community-wiki answers of a minimum length are eligible