Formatar SSD 1Tb fresquinho: afinal quantas e quais partições?

The kernel dump operation is done right after a panic on a raw device. This raw device is either the swap area or a dedicated partition.

The savecore operation is done when the OS reboots and indeed create files on a filesystem.

The latter operation doesn't happen if the former one didn't so I still assert either a swap or a dedicated partition is required for the kernel crash dump to succeed [or] a filesystem.


[...]


The dump device is where the kernel will write the crash dump when it encounters a panic condition. This is always a raw area on disk. Remember that a panic means that the kernel has malfunctioned. No one would want a malfunctioning kernel to attempt to write a very large file into a file system. The very reason for a panic is to stop I/O to filesystems before they are more badly damaged.

The savecore program runs during a reboot (but after fsck's and thus we are sure that the filesystem can be used). Usually the dump device is not dedicated but is a swap area so we want to crash dumps before swap is needed. savecore reads from the dump area and write to the file system. We clearly need to save an image of the damaged kernel before we reboot. After a reboot, the broken kernel is gone.

[...]

One of the reasons why an option was added to dumpadm to allow not using the swap space was precisely swap space not being allocated large enough for crash dump to succeed.

To properly size the swap area, there are two factors to take into account.

1: at the peak virtual memory consumption, all memory reservations need to be backed by either RAM or swap area pages. This doesn't means any activity will necessarily occur on the swap, just the pages have to be there. Solaris isn't doing lazy memory allocation like the Linux kernel does. The free virtual memory can be easily monitored and it is easy to increase its size by adding new swap devices or swap files, so this first point isn't that much of a problem.

2: The default Solaris configuration is to have crash dumps to be written on the swap. Many (but not 99.9% of) Solaris installations are using this standard configuration. In the old Solaris 2.x days, the swap area was very often larger than the RAM size because the (already obsolete) rule of thumb x2 was still followed. There was no risk for the crash dump not to fit. Nowadays, it is not uncommon to have the RAM much larger than the swap. That means there is a risk to miss the invaluable data stored there should a panic happen. I'm not telling this risk cannot be overcome, I'm telling many people are unaware of this risk because it is not common to monitor the kernel size and even less common to check if it would fit in the dump device should a panic occur.

[..]

Swap doesn't slows you down, lack of RAM slows you down when swap is available and crashes your application when no swap is there. I strongly prefer my applications to slow down and the OS to paginate than seeing random applications crashing because of malloc failures.