anarazel

65 ms·

691.
by anarazel 8y ago
Well, the comment was in reply to somebody asking for the postgres feature to be locked down similar to MSSQL's - and I think the "FUD" spreading commenter just was trying to explain that it's already similarly locked do
692.
by anarazel 8y ago
Because enough users asked for them - there was a long debate about that. It's less to prevent those users from intentionally gaining superuser, and more to make it a bit harder to unintentionally have trusted users shoot themselves in
693.
by anarazel 8y ago
The quote's from me, and even in the quoted part explicitly says "if you have a privileged account". And an admin user you just need to enable it, and it works without even the proxy user configured. Please explain what&#x2
694.
by anarazel 8y ago
Maybe I'm just a grump, but: 5% is pretty low number for training people up. Perhaps that's just the "hit the books on your own, but during work time" portion? If it includes other things (talking people though a project
695.
by anarazel 8y ago
FWIW, in hindsight, this never should have been called pluggable storage, but rather pluggable table access methods, or at least pluggable table storage. But it was was in the subject on the first thread this patchset originated in (by Alv
696.
by anarazel 8y ago
I think it's a significant danger. But on the other hand our current table storage has some architectural issues that are hard to fix in an upward compatible manner (causing a number of issues around vacuum and write amplification du
697.
by anarazel 8y ago
Uh, thanks, I guess? I'm Andres, the committer mentioned. Perhaps next time speak up before everything's committed?
698.
by anarazel 8y ago
Note that that often will be doable just as an expression index. Say CREATE INDEX foo ON bar(my_function(cola, colb)); which then can answer questions like SELECT * FROM bar WHERE my_function(cola, colb) = 'zaphod'; if my_function
699.
by anarazel 8y ago
As a vegetarian of ~20 years, I kinda find them too meat like. I constantly wonder whether I accidentally got the wrong patty. I guess that's largely because I've only dim memories of the real thing. Am I the only one in that?
700.
by anarazel 8y ago
> The problem is that 2MB pages are too much to duplicate for a single byte change. Splitting up such pages into their 4kb counterparts when dirtying a COWed huge page isn't particularly crazy. I don't think there's much r
701.
by anarazel 8y ago
IOW, a bad persistence design in redis currently interacts with a suboptimal COW-after-fork implementation in linux.
702.
by anarazel 8y ago
Is that still true with recent kernels? Around postgres we'd huge problems with earlier kernels with enabled THP, but with recent kernels its gotten to be a pretty small penalty.
703.
by anarazel 8y ago
> Amazon.com as well. In the beginning one of their biggest competitive advantages was that they didn't charge sales tax. Wasn't that basically the law of the land (until this year, due to Wayfair), based on the Quill SCOTUS de
704.
by anarazel 8y ago
Hear hear.
705.
by anarazel 8y ago
Huge numbers of postgres installations are on small machines and/or machines shared with other software. And whether you want a large work_mem hugely depends on whether you want to run an OLTP workload with a few hundred connections,
706.
by anarazel 8y ago
Flying business I can reasonably work during the flight, and for overnight flights I can sleep. Which reduces the productivity loss due to lack of sleep (I arrive with a few hours of sleep) and jetlag (adjustment happens quicker than when I
707.
by anarazel 8y ago
Nope. There's a lot of legislation passed in one chamber just to die in the other. I don't know how many times there have been bills to repeal the ACA in the house, for example, that never got voted upon in the Senate.
708.
by anarazel 8y ago
Passing a bill in the house doesn't have much of a forcing function on getting a bill voted upon in the Senate. I doubt McConnell will bring it up for a vote. It's perhaps possible to attach it to some must pass legislation, in co
709.
by anarazel 8y ago
A lot of formerly YIMBY people become very NIMBY as soon as they own property. I've seen that with a number of people succeeding in the startup game, for example. I don't think the lines are as clear as you make them to be.
710.
by anarazel 8y ago
It just seems like it has nothing to do with the article. Upon rereading they even have his name in there: Mr. Murry, who is an adviser for Representative Kevin McCarthy of California, raised doubts over the weekend about whether that
711.
by anarazel 8y ago
The aide is actually not anonymous if you follow the story a bit. If you follow the links in the story you can easily figure it out. I think it's good to not expose career staffers to the full public public wrath without need. Making i
712.
by anarazel 8y ago
They're actively working on it. There's been a new revisions posted over the last few days, responding to review from a week or two ago. https://www.postgresql.org/message-id/a3be6a7a-77d3-0e88-4f9... https:
713.
by anarazel 8y ago
> And this is actually what you want from a performance point of view: Reduce the number of roundtrips and keep your transactions as short as possible. Uhm, obviously. But it turns out there's other constraints too. Like whether opt
714.
by anarazel 8y ago
> I don't know how MySQL behaves in this regard, but in Postgres the number of transactions the system can run concurrently is roughly limited by the number of CPUs you have available. At some point adding more sessions just means t
715.
by anarazel 8y ago
FWIW, I don't think CSN actually gets us that far, because the data structures to map xids to CSNs aren't that cheap to maintain either (which is needed for MVCC visibility tests).
716.
by anarazel 8y ago
> Interesting, but surely snapshots are only needed for connections with active transactions? Right, but the problem is that the cost to compute them essentially scales with the number of active connections (and to some degree with the
717.
by anarazel 8y ago
> MySQL's ability to handle many thousands of idle connections without breaking a sweat actually solves this problem, and does so in a very convenient way, particularly in a micro-services architecture, and I really wish Postgres su
718.
by anarazel 8y ago
Not gonna happen that soon, sorry. There's a few easy enough changes to improve PG connection scaling considerably, but to get to that count more invasive changes are needed.
719.
by anarazel 8y ago
And in PG 12 we've removed the magic oid columns (as in WITH OIDS), although you can still create one manually (but there'll not be any values assigned automatically).
720.
by anarazel 8y ago
I don't see how you'd describe postgres validly as a key-value store at the storage level / internally. It's a pretty traditional row-store (without support for index oriented tables), i.e. all columns of a table are sto

More ›