Huge pages are what they are called- huge pages. Normal page size by default is 4kB on x86 and x86_64. Huge pages are 4MB in size on x86, 2MB on x86_64, and 256MB on IA64
What is a page? Memory is allocated to processes in pages.
A page, memory page, or virtual page is a fixed-length contiguous block of virtual memory, described by a single entry in the page table. It is the smallest unit of data for memory management in a virtual memory operating system
What is the benefit of huge? Continue reading
…and how to identify the filesystem type if there is one.
/dev/sda1 has an ext4 filesystem
file -s /dev/sda1
/dev/sda1: Linux rev 1.0 ext4 filesystem data (needs journal recovery) (extents) (huge files)
If it is a logical volume
file -s /dev/mapper/vg-log_lv
/dev/mapper/vg-log_lv: symbolic link to `../dm-8'
file -s /dev/dm-8
/dev/dm-8: SGI XFS filesystem data (blksz 4096, inosz 256, v2 dirs)
“They had an under-performing employee that everyone liked & no-one wanted to fire, so they moved him through different roles to find something that suited him. After moving him through the entire company with negative results, they finally conceded they had to fire him. The punch line: the employee flourished as soon as he was fired – his dream had been to join the ministry, but he’d worried about quitting a stable job to do so.”
I built new hosts which were added to an existing cluster. After taking them out of maintenance mode, I cannot vMotion VMs to the new hosts. I followed this KB article and this blog post. Nothing worked. I was on the verge of resetting one host.
vim-cmd vimsvc/task_list showed several tasks like below, it didn’t seem right.
I thought of checking jumbo frame which turned out to the culprit. This cluster uses VSS. I changed the MTU to 9000 for VMkernel NAS interface but forgot to do so at the vSwitch level which is 1500 by default. After changing the MTU at the vSwitch level, it worked immediately.