chucky_z

Did you read this?

https://www.postgresql.org/docs/current/static/high-availabi...

I'm not sure what more you can ask from documentation.

Does MySQL document Vitess, Galera, MaxScale, etc...?

What replication issues are you referring too? I never once had a problem with pgsql's replication across 8.1->9.5.

It would randomly die, but that was always either my fault or the applications fault, never pgsql itself.

The lack of master-master seems to be the big thing everyone mentions, but PostgresXL is currently in a usable-in-production state.

online schema change is absolutely riddled with crippling bugs if you use foreign keys. Just a heads up, we've had to basically take everything it does and make an in-house version. The idea and execution are great, when it actually works.

From my experience, the problem is not how does any NoSQL solution stand up vs. any other NoSQL solution. The question to ask is "What caveats affect me the least?" and then go with that NoSQL solution. In the case of MongoDB, the biggest issue (at least in the past) is that the caveats simply aren't known well, or at all!

Compare this to something like Elasticsearch, which is well-known (and well-documented) to have data loss in certain situations.

I will argue _against_ MongoDB all day, but there is most certainly a time and place for it. If you have a real firehose of data, don't care too much about the accurateness, and really need good speed... MongoDB is not an awful choice.

Regarding variable precedence, it's pretty simple to grok in 2.x+, and well-documented now (it was a serious nightmare, and a very valid complaint... but isn't as much anymore).

For sharing common group_vars, what I do is my inventory passes an environment-aware variable which I include in the playbook by doing like:

`- include: "env_vars/{{ environment }}/main.yml"`

I feel I sidestepped some of the AWS modules issues simply by using Terraform. I still build all my images in Packer using the Ansible builder though.

That's roughly 4k qps. Assuming it's a lightweight API a single server isn't out of the question. Nginx will handle 20k qps until the end of time on a tiny box serving static content.

The article only mentions their app. I try to install as few apps as possible for personal reasons.

Their mobile web UX is turning sour, fast. (I primarily used the messenger function, so my view is very skewed the)

I don't have any strong things to point at other than the messenger function, but it just feels far worse than it did a year ago.

My response to the comment for triggers and PGSQL would be: Postgres tries its best to stop you shooting yourself in the foot.

Similar to the top comment, all the real problems I've ever encountered with postgres (heck, all major RDBMS's for that matter) come from certain areas, mainly triggers.

I use Ansible to do all of the CIS recommendations. Took about a day and following along with the CIS guide to do it.

I was simply disagreeing with the title of "SELinux is beyond saving." SELinux doesn't need saving for those who have a strong use-case. There are some great lesser alternatives like AppArmor, grsec, TOMOYO, and AKARI if SELinux can't work for you. :)

May I disagree partially, using this authors exact words?

"... Unless you are a high risk target, spending almost any time beating SELinux into shape on your machine is a bad tradeoff and pretty much a waste ..." -- https://utcc.utoronto.ca/~cks/space/blog/linux/SELinuxToxicM...

If you are a high risk target, SELinux is a great option.

Also, to refer to the partial disagreement, here is my exact feeling...

"I'm a security aware sysadmin and yet yesterday I casually admitted that I made less-secure choices because the really secure option was too annoying and potentially inconvenient. In fact this is not the only case where I make this tradeoff, picking a less secure but more convenient option..." -- https://utcc.utoronto.ca/~cks/space/blog/tech/SecurityNotImp...

Everyone should try PostgreSQL with hstore (and JSONB now, too!).

This is a key/value store inside an RDBMS that just works, and it works great!

I converted a crappy sloppy super messy 1000+ column main table in a ~800GB database to use hstore, it was, in real world benchmarks, between 7x and 10,000x (yes, really, ten thousand times) faster.

The CEO of the company who had a technical say in everything, and was very proud of his schema "wasn't excited" and it never happened in any production instance.

I've left since then, and the company has made very little advancement, especially when it comes to their database.

Really, just use hstore. Try it out. The syntax is goofy, but... I mean, SQL itself is a little bit goofy, right?

Postgres scales great up to 256gb, at least with 9.4. After that it'll use it, but there's no real benefit. I don't know about MySQL. SQL Server scales linearly with memory even up to and past the 1TB point. I did encounter some NUMA node spanning speed issues, but numactl tuning fixed that.

I setup a handful of pgsql and Windows servers around this size. SQL Server at the time scaled better with memory. Pgsql never really got faster after a certain point, but with a lot of cores it handled tons of connections gracefully.

I've confirmed this... It's an opt-out proxy in this situation (can be setup as opt-in).

https://crypto.stanford.edu/flashproxy/

> There is an options page[1] (the same page you get by clicking on the badge) with which users can choose whether they want to be a proxy. By default, if a user has not made a choice, they will be a proxy. If you want only people who have explicitly clicked "yes" to be a proxy, add the cookierequired parameter. If a user has selected "no," they will never be a proxy, regardless of the presence of cookierequired.

[1] https://crypto.stanford.edu/flashproxy/options.html

Interesting you mention awk. Do you use that professionally? For string manipulation I use almost exclusively pipe output to bash then format/grok with awk/sed, and wish that more (any?) languages implemented them fully!

A real announcement (unicode support) in an April Fool's package (doing the announcement only mentioning emoji).

LukeShu nailed it with their reply. :)

Interesting there seems to be a shift in the 0-2 year job length position. I wonder if this is because the common trend is to jump ship at 2 yrs, so folks are going out job hunting, and getting a competitive raise at their current position with the job offer in hand?

We have a backend position (primarily PHP) open at our company (am I allowed to pimp that outside the who's hiring thread?). It's been open for ~6 months or so now, and haven't had a single female applicant. Not, we haven't had any female interviews; we haven't had any female _applicants_.

Our company is almost 50/50 for male/female, with our engineering team being 100% male. At the last company I worked for in the Tampa area of FL, we had one female developer who was MTF, and we did get one female applicant to an open position, who wasn't hired because she was deemed too junior. This was for a junior position, and I think the CEO was just being unknowingly sexist.

People pay for RHEL, SUSE, Windows, Mongo, MySQL, etc.. support even though they may be part of a company perfectly capable of resolving any issue with the software. You get more than just basic support with all of these contracts. =)

Our indexes are probably very small (about 40gb total) but I've found if you value speed and consistency over all else.... just run a single, 1 shard node and have a fallback in case of random errors. I just do major version upgrades and rebuild the indexes, no real maintenance.

I should add that all "hot" data fits in 24g of memory.

As a consumer of AWS's services, every time I start to think "I could really use this service..." or I have a "If AWS had this..." thought, they seem to release something that addresses this concern. If they added multicast within VPC I would be the happiest lil' AWS camper.

Isn't the whole point of a devops individual/team to reduce friction to "... break down barriers, increase automation, bring collaboration and iterate."?

I understand that there should be company buy-in and individuals could handle this themselves but if there's no clear start/end point there could (will) be a huge waste of man hours and potentially $$.

So maybe a subtitle for this article could be "But it may help."

The only thing I've found strongly lacking in bash is math. You can `| bc` all you want, but it still seems slow compared to other languages. I typically do all my math in python (typically through pipes... heh), and everything else in bash.

Check this out, I think you'll be impressed!

https://www.gnu.org/software/gawk/manual/gawkinet/gawkinet.h...

(If you want it to fully function you'll need bits of code starting here -- https://www.gnu.org/software/gawk/manual/gawkinet/gawkinet.h...)

Actually, the insane amount of advertising and media groups on Facebook have pushed me to Twitter, and it's very refreshing. I'll be moving away from everything but FB Messenger (imo their best product).

Correct.

I did this migration recently and we're spending about 1.75 cents per user. We could do it for cheaper, but we've recently had some issues that were absolutely trivial to resolve with AWS, that would have been very difficult with our previous hosting provider.

I saw the little text box at the bottom jiggle at me and immediately closed the page.

#1 PROTIP to a good UI. Don't jiggle anything, anywhere, ever, for any reason.

400k emails isn't that many though. :(