A
Anchor Blog
anchor.com.au · Company blog · on HN since 2009
Anchor is an Australian sysadmin company whose blog covers Linux, storage, GitHub's architecture, rsync, and software failures.
What HN saysHN commenters recommend it to sysadmins for good technical content and amusing criticism of software, and cite its posts on SSH ControlMaster and database issues.
29HN people shared it
13linked it in comments
66stories submitted
618story points
6 years agolast shared
2009people per year2026
Most-shared posts
- 1GitHub's Architecture, before and after migration3 people · 80 points · first shared 2009 · 14 comments on HN
- 2SSH ControlMaster: The Good, the Bad, the Ugly (2010)3 people · 2 points · first shared 2013 · discussion on HN
- 3Out-Tridging Tridge by improving rsync2 people · 54 points · first shared 2013 · 13 comments on HN
- 4Load balancing at Github: Why ldirectord?2 people · 27 points · first shared 2009 · 5 comments on HN
- 5Why does Percona cause SSL issues in Postfix?2 people · 3 points · first shared 2012 · discussion on HN
- 6A review of the btrfs filesystem2 people · 1 points · first shared 2013 · discussion on HN
- 7Awesome but commonly unknown Linux Commands1 people · 167 points · first shared 2011 · 87 comments on HN
- 8Hunting deep-seeded kernel bugs1 people · 75 points · first shared 2012 · 16 comments on HN
- 9MySQL vs. PostgreSQL comparison by Anchor (github's migration team)1 people · 44 points · first shared 2009 · 42 comments on HN
- 10Inquisitio interrupta - squashing a MySQL connection-dropping bug1 people · 24 points · first shared 2012 · 4 comments on HN
- 11Putting OpenBSD in the cloud1 people · 9 points · first shared 2015 · discussion on HN
- 128 reasons that a VPS is more awesome than you realise1 people · 8 points · first shared 2012 · 21 comments on HN
- 13GitHub fork their sysadmins1 people · 8 points · first shared 2011 · discussion on HN
- 14Web developers: The Internet is trying to kill you1 people · 7 points · first shared 2011 · 4 comments on HN
- 15Github: Speed Matters1 people · 7 points · first shared 2009 · 3 comments on HN
What HN people wrote when they linked it
zelphirkalt · 2021-05-02
> Why this isn’t the default I don’t know, [...]
I searched a little and found the following blog, which claims, that there are issues with it, when you do heavy data transfers, for example via many rsyncs at the same time:
https://www.anchor.com.au/blog/2010/02/ssh-controlmaster-the-good-the-…
I did not test it myself. This would seem like an appropriate reason to not make it the default.
kezza · 2012-03-28
For what it is worth, It was reported to RedHat, but our experiences with them have proved to be fairly.. hmm.. lackluster.
We've talked about it before http://www.anchor.com.au/blog/2009/02/the-value-of-commercial-software…
sciurus · 2011-08-10
Anchor's blog is worth subscribing to if you're a sysadmin. The technical content is good, and they can be merciless yet amusing when writing about software that fail to meet their standards of good taste (see http://www.anchor.com.au/blog/tag/fail/).
datums · 2009-09-28
I was looking at http://www.anchor.com.au/blog/wp-content/uploads/2009/09/GitHubStorage… which shows file server B as a failover, traffic coming into the active LB.
Does the active LB then have a pool which includes other file server B services ?
hyc_symas · 2019-01-14
Here ya go. Quite critical of Kyoto Cabinet.
https://www.anchor.com.au/blog/2013/05/second-strike-with-lightning/
shizcakes · 2013-09-05
If you're a heavy SSH user, using multiplexing in this manner can have negative consequences [1]. Downsides include having all your multiplexed connections exiting if the master exits!
[1] http://www.anchor.com.au/blog/2010/02/ssh-controlmaster-the-good-the-b…
tjoff · 2013-07-23
It's a bit different with a file system that still undergoes heavy development of basic features...
http://www.anchor.com.au/blog/2013/04/the-btrfs-backup-experiment/
Just don't use btrfs where a failure is going to hurt.
nasalgoat · 2013-06-08
You might be interested in this:
http://www.anchor.com.au/blog/2013/04/redis-rethought-exciting-extreme…
krembo · 2012-08-16
Even if everything you say is true, taking a look into your VPN prices pages, made me almost fall off my chair
http://www.anchor.com.au/servers/
vimalg2 · 2011-10-04
Reminds me of the last Github post I'd read about their load-balancer setup from their server guy(2009):
http://www.anchor.com.au/blog/2009/10/load-balancing-at-github-why-ldi…
Willy Tarreau (author of HAproxy) had sparked a nice discussion in the comments section, that time.
timf · 2011-08-30
You may have heard of Anchor because of GitHub: http://www.anchor.com.au/blog/2009/09/github-designing-success/
jameskilton · 2011-01-30
Not sure how this is relevant. Github isn't hosted on any cloud.
http://www.anchor.com.au/blog/2009/09/github-designing-success/
peterwwillis · 2015-07-22
This is also sometimes called 'checkpointing', and has been a long-studied feature necessary for distributed processing systems. There's a lot of ways to skin this cat. Kernel modules, application libraries, userland tools, and features to support various kinds of applications (single-threaded, multi-threaded, preserving TCP connections, etc). Find the one that fits your use best.
Descriptions of how checkpointing works: https://en.wikipedia.org/wiki/Application_checkpointing http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.117.4869&… http://www.anchor.com.au/blog/2013/02/overview-of-checkpoint-and-resto… https://lwn.net/Articles/478111/ https://www.usenix.org/legacy/event/usenix01/freenix01/full_papers/die…
Individual solutions: https://ckpt.wiki.kernel.org/index.php/Main_Page https://code.google.com/p/cryopid/ http://criu.org/Main_Page http://crd.lbl.gov/departments/computer-science/CLaSS/research/BLCR/
An almost-full list of solutions here: http://checkpointing.org/