Open Source

replacing an existing zroot with a proper zroot

I wrote this post late one night while pondering how to solve this problem. I may have solved it with much less effort. I want to copy a zroot from old drives to new drives. The new drives are in a test box of mine. Once the new drives are configured, I will replace the existing mirror with them. Not shown here, I have already installed FreeBSD 12.1 on the new drives. I […]

replacing an existing zroot with a proper zroot Read More »

ZFS: adding a drive back into the zpool

Today I was updating some servers. One of them was rebooted three times. On the third time, one of the drives went missing. This is how I tracked down which drive, and which slot in the server, and fixed it. I’m writing it mostly so I can remember how to flash the light on the drive. There is also an off-by-one issue to avoid. In this post: FreeBSD 12.1 sesutil ZFS Supermicro 846

ZFS: adding a drive back into the zpool Read More »

using zfs rollback for cache clearing

I’m in the final stages of the FreshPorts packages project. One of the last tasks is clearing the packages cache from disk when new package information is loaded into the database. See also: Clearing the packages cache ZFS when you’re not doing ZFS Several of the configuration items have been learned from putting my poudriere instance into a jail. In this post: FreeBSD 12.1 py37-iocage-1.2_5 If I was to do this cache clearing

using zfs rollback for cache clearing Read More »

WARNING: failed precmd routine for mysql

Last night the MySQL server instance in my dbclone jail stopped working. This is the jail into which copies of all my databases eventually go for test restores and backups. This morning I tried started it: [dan@dbclone:~] $ sudo service mysql-server start /usr/local/etc/rc.d/mysql-server: WARNING: failed precmd routine for mysql My initial search suggested removing /var/db/mysql (this will lose all your data). Since this was a testing server, I tried it. No success. I

WARNING: failed precmd routine for mysql Read More »

bectl details

This is just a placeholder for me to find this later. I was cleaning up some old snapshots. I’m not sure I should have removed those snapshots and I’m saving this here for next time I use bectl. [dan@slocum:~] $ grep -v autosnap ~/tmp/snapshots/snapshots zroot/bootenv/default@2020-01-11-18:16:51 zroot/bootenv/default/tmp@2020-01-11-18:16:51 zroot/bootenv/default/usr@2020-01-11-18:16:51 zroot/bootenv/default/usr/local@2020-01-11-18:16:51 zroot/bootenv/default/usr/obj@2020-01-11-18:16:51 zroot/bootenv/default/usr/src@2020-01-11-18:16:51 zroot/bootenv/default/var@2020-01-11-18:16:51 zroot/bootenv/default/var/audit@2020-01-11-18:16:51 zroot/bootenv/default/var/empty@2020-01-11-18:16:51 zroot/bootenv/default/var/log@2020-01-11-18:16:51 zroot/bootenv/default/var/tmp@2020-01-11-18:16:51 [dan@slocum:~] $ sudo zfs destroy zroot/bootenv/default/var/tmp@2020-01-11-18:16:51 [dan@slocum:~] $ sudo zfs destroy zroot/bootenv/default/var/log@2020-01-11-18:16:51 [dan@slocum:~] $ sudo zfs destroy zroot/bootenv/default/var/empty@2020-01-11-18:16:51

bectl details Read More »

The backups have been wrong for about 97 days

I backup my databases daily. They get dumped to a disk file and copied somewhere else. Then they are backed up into a file system. Once a month they get copied to tape. Yesterday I upgraded a PostgreSQL database server, and it went well. Overnight, a backup failed. This is the story of how that led to me discovering my backups have been at risk for nearly 100 days. The error This morning

The backups have been wrong for about 97 days Read More »

Creating a ZFS dataset for testing iocage within a jail

Be warned, this failed. I’m stalled and I have not completed this. I’m going to do jails within a jail. I already do that with poudriere in a jail but here I want to test an older version of iocage before upgrading my current jail hosts to a newer version. In this post: FreeBSD 12.1 py36-iocage-1.2_3 py36-iocage-1.2_4 This post includes my errors and mistakes. Perhaps you should proceed carefully and read it all

Creating a ZFS dataset for testing iocage within a jail Read More »

What files installed by this package have been modified post-install?

You’ve seen it. A package gets installed. Some shell scripts are included. They get modified. It happens. But how do you know what has changed? I know there is a tool in pkg for this. I know there is a periodic script which uses it. Let’s go looking. In this post: FreeBSD 12.1 periodic $ cd /usr/local/etc/periodic/ $ find . | grep checksum ./security/460.pkg-checksum There it is! Looking inside, I found pkg check.

What files installed by this package have been modified post-install? Read More »

Migrating FreshPorts from one db server to another

FreshPorts runs on a FreeBSD server which hosts multiple jails. Two of these jails run PostgreSQL server. When upgrading from one version of PostgreSQL to another, we run pg_dump in the new jail, and load the backup into that database server. I’m writing this blog post to keep track of this procedure so I do not have to remember it each time. take website offline sudo mv mv offline.conf.disabled offline.conf && \ sudo

Migrating FreshPorts from one db server to another Read More »

pkg: vulnxml parsing error: no element found

Today I found this annoying situation on FreeBSD 12.1 in a FreeBSD 12.0 jail (neither of which are directly relevant to the problem at hand). [dan@serpico:~] $ sudo pkg audit -F vulnxml file up-to-date pkg: vulnxml parsing error: no element found pkg: cannot process vulnxml After a bit if thinking, I figured the vulnxml file was corrupt. I guessed it might be in /var/db/pkg: [dan@serpico:/var/db/pkg] $ ls -l total 5226 -rw-r–r– 1 root

pkg: vulnxml parsing error: no element found Read More »

Scroll to Top