Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

AirBNB Opensources Chronos, a Cron Replacement

Unknown Lamer posted about a year and a half ago | from the end-of-time dept.

Open Source 72

First time accepted submitter victorhooi writes "AirBNB has open-sourced Chronos- a scheduler built around Apache Mesos (a cluster manager). The scheduler is distributed and fault-tolerant, and allows specifying jobs in ISO8601 repeating notation, as well as creating dependent jobs. There's also a snazzy web interface to track and manage jobs, as well as a RESTful API." It's under the Apache License as seems to be the fashion with businesses releasing software nowadays. It looks like it might be useful if you have to manage a lot of machines with interconnected recurring processes; I know I wish this had existed a few years ago.

cancel ×

72 comments

Sorry! There are no comments related to the filter you selected.

Unnecessary. (-1)

Anonymous Coward | about a year and a half ago | (#43210389)

Cron does everything I need it to do. I don't see a need to complicate things by duplicating functionality.

Re:Unnecessary. (5, Insightful)

viperidaenz (2515578) | about a year and a half ago | (#43210431)

"distributed and fault-tolerant" and "dependent jobs"
All things cron doesn't do.

Re:Unnecessary. (3, Insightful)

kangsterizer (1698322) | about a year and a half ago | (#43211063)

Which is exactly why its not a cron replacement. Anybody who think this == cron had NO clue of what they were doing when they were using cron, and still doesn't.

... cron on steroid ? (1)

Taco Cowboy (5327) | about a year and a half ago | (#43211171)

Which is exactly why its not a cron replacement.

Would "Cron on Steroid" satisfy you?

Re:... cron on steroid ? (1)

gayak (745124) | about a year and a half ago | (#43211955)

No, but this seems to be aimed more at Control-M and other scheduling 'frameworks'. Not that these features are enough to challenge CTM, but it's still targeting that one more than it's targeting cron.

Re:Unnecessary. (1)

MikeBabcock (65886) | about a year and a half ago | (#43216617)

This is basic set theory.

If ProductB does *at least* everything ProductA does then it can be a ProductA replacement.

If ProductB does more things, that's not relevant to its use as a replacement.

Only if ProductA has features that ProductB cannot duplicate does ProductB fail to be a possible replacement.

So what features does cron have that this does not?

Re:Unnecessary. (1)

roman_mir (125474) | about a year and a half ago | (#43211645)

I know, chronos is so great, it will launch your app even if the machine is down itself! It'll supply the power that comes stored in a special file to do it.

Re:Unnecessary. (2)

interval1066 (668936) | about a year and a half ago | (#43210655)

Yeah. Why use squid when you can just stuff client requests in a hash table, too...

Re:Unnecessary. (1)

dintech (998802) | about a year and a half ago | (#43211909)

Yes, 640k ought to be enough for anyone.

Re:Unnecessary. (0)

Anonymous Coward | about a year and a half ago | (#43213923)

64k was enough for everyone as Steve Jobs knew so 640k out to be more than enough.

Re:Unnecessary. (1)

jellomizer (103300) | about a year and a half ago | (#43216901)

Because anything new is bad.

Keep it simple, stupid (4, Insightful)

manu0601 (2221348) | about a year and a half ago | (#43210403)

This is not a replacement for cron. On an isolated machine, it would be foolish to trade cron for such a complicated beast. On many nodes, I understand it has benefits.

Re:Keep it simple, stupid (3, Interesting)

Anonymous Coward | about a year and a half ago | (#43210573)

Here are some even simpler and securer alternatives:

http://code.dogmap.org/runwhen/
"runwhen is a set of utilities for running commands at particular times. With these tools, you can perform calculations on time values in various ways, and use those calculated times to determine how long a process should sleep before performing some task."

http://ohse.de/uwe/uschedule.html

http://untroubled.org/bcron/
"This is bcron, a new cron system designed with secure operations in
mind. To do this, the system is divided into several seperate programs,
each responsible for a seperate task, with strictly controlled
communications between them. The user interface is a drop-in
replacement for similar systems (such as vixie-cron), but the internals
differ greatly."

http://www.superscript.com/trigger/index.html
"The trigger package contains tools for running programs on demand, invoked by external programs via data written to a fifo."

Re:Keep it simple, stupid (1)

jockm (233372) | about a year and a half ago | (#43212907)

I'll give you simpler (potentially), but could you explain why you say these are more secure?

Re:Keep it simple, stupid (0)

Anonymous Coward | about a year and a half ago | (#43215355)

I'll give you simpler (potentially), but could you explain why you say these are more secure?

These folow the KISS principle and are designed with security in mind from the start.

Re:Keep it simple, stupid (1)

jockm (233372) | about a year and a half ago | (#43223319)

KISS is no guarantee of security. In what way are they simpler and thus more secure. While complexity introduces more opportunity for insecurity, inverse is not necessarily true. If you are just asserting that simple software is somehow better then you need to back that up.

The statement made was that the examples listed above were more secure, which implies familiarity with them. If you are just talking in generalities that isn't actually very useful.

Re:Keep it simple, stupid (0)

Anonymous Coward | about a year and a half ago | (#43225689)

If you'd look at bcron's implementation page, and others' you'd see why. Don't take my word for it.

Re:Keep it simple, stupid (1)

jockm (233372) | about a year and a half ago | (#43234289)

Please remember the first rule of skepticism, the person to make the claim is the one who should justify it. Simply telling me to go look at bcron's implementation page doesn't do that. Especially since four different schedules were mentioned.

You (I presume you) are telling me that Chronos is insecure, you need to actually prove that. What are the vulnerabilities, or are you just speaking in vague generalities based on some philosophical belief?

Re:Keep it simple, stupid (0)

Anonymous Coward | about a year and a half ago | (#43237401)

Please remember the first rule of skepticism, the person to make the claim is the one who should justify it. Simply telling me to go look at bcron's implementation page doesn't do that. Especially since four different schedules were mentioned.

You (I presume you) are telling me that Chronos is insecure, you need to actually prove that. What are the vulnerabilities, or are you just speaking in vague generalities based on some philosophical belief?

You don't understand, the software I mentioned are alternatives to cron, just like Chronos. I don't know much or care about Chronos, as my cron needs are already met.

Re:Keep it simple, stupid (1)

jockm (233372) | about a year and a half ago | (#43238393)

OK Fair enough, though you could have been more clear. That being said, you still haven't actually justified your statement why any of the things you mentioned are more secure than cron.

Re:Keep it simple, stupid (0)

Anonymous Coward | about a year and a half ago | (#43261027)

These were designed as more secure alternatives to cron, because the authors were dissatisfied with cron's security.

Nothing to do with OpenGL (1)

tepples (727027) | about a year and a half ago | (#43210441)

I wonder what the Khronos Group would have to say about the name of this project.

Re:Nothing to do with OpenGL (1)

bill_mcgonigle (4333) | about a year and a half ago | (#43210549)

I wonder what the Khronos Group would have to say about the name of this project.

They might snicker at the latinization of the name of the Greek god?

Where do I start ... (4, Interesting)

recrudescence (1383489) | about a year and a half ago | (#43210685)

a) Chronos is actually correct (to the extent that the most accepted transliteration for the greek letter chi is 'ch' rather than 'kh') and means 'time'.
b) If anything, it's actually the Khronos group which should be cowering in shame, since they are misspelling the name Kronos.
c) Latin doesn't even have a 'ch' diphthong, except when transliterating Greek words (http://en.wikipedia.org/wiki/Ch_%28digraph%29#Latin)
d) The latinization of Kronos would have been Cronus, not Chronos.
e) Strictly speaking, Kronos is a Titan, not a Greek God (except in the looser definition of Titans as deities in general)

Fail.

Re:Where do I start ... (0)

Anonymous Coward | about a year and a half ago | (#43212485)

...
c) Latin doesn't even have a 'ch' diphthong, except when transliterating Greek words (http://en.wikipedia.org/wiki/Ch_%28digraph%29#Latin) ...
Fail.

'Ch' is not a diphtong, it's a digraph. The collocation of two sounds, not the collocation of 2 characters. Even the wikipedia article you quote said it's a digraph.

Re:Where do I start ... (0)

Anonymous Coward | about a year and a half ago | (#43212929)

But isn't it a single sound ? Specifically, the same sound as the X in TeX should be ? From what I gather Spanish, Russian and Hebrew all has it too. But in English there's just no standard notation for it so weird things happen like when it's from Greek -> Latin -> English it's the one thing, but when it's directly from Greek it's the other way around (K \ Kh \ Ch).
Then you have the Hebrew, Yiddish and Russian words using H alone or with some odd vowel. And Spanish uses X for some reason.
And then you get into the Asian transliterations which don't even try to match sounds to letter so things just completely break there.

(TLDR)
Simply put, the problem is the English writing system. And calling it a system is already suggesting too much.

Re:Where do I start ... (1)

recrudescence (1383489) | about a year and a half ago | (#43253789)

Apologies; in Greek grammar the same effect is called a diphthong. Nice to know of the difference in English grammar though :)

Re:Nothing to do with OpenGL (1)

flimflammer (956759) | about a year and a half ago | (#43211313)

Nothing.

I think I speak for everyone when I say ... (-1, Redundant)

OhANameWhatName (2688401) | about a year and a half ago | (#43210449)

.. yawn.

Chronos, and Apache License thoughts (2, Informative)

ciurana (2603) | about a year and a half ago | (#43210577)

Chronos looks very yummy. Over the years I've deployed a number of schedulers (launchd on OS X and Quartz come to mind) but cron always comes back because it's so available and flexible. While it has many shortcomings, it's reliable and easy to grasp. Chronos, with the ISO 8061 job scheduling syntax will have an edge over the nasty mess of launchd, and the cron-like extensions and idiosyncrasies in Quartz. The first glance at the GitHub pull shows clean code. I'm looking forward to taking it through its paces on OS X and Linux.

Unknown Lamer wrote:
> It's under the Apache License as seems to be the fashion with businesses releasing software nowadays.

It's not a matter of fashion, it's a practical reality. No sane business wants to be the who defends the GPL in court. It'll be expensive and messy, and if the result goes against GNU/GPL "accepted wisdom", it will be a PR nightmare. The Apache License strikes a good balance between permissiveness and restrictions: less restrictive than the *GPL, less permissive than BSD or MIT. I advise various companies (startups, public, etc.) and venture funds on this regard. We recently advised someone using mongoDB (GPL3) to ensure that they built a very flexible abstraction layer between the app and the database that, by design, would allow swapping to something different (e.g. Cassandra, CouchDB, etc.) with a less restrictive license than *GPL and with similar characteristics. That single item, mongoDB's license, could be the deciding factor between getting funding/being acquired or not.

Cheers!

pr3d

Re:Chronos, and Apache License thoughts (5, Insightful)

Anonymous Coward | about a year and a half ago | (#43210761)

It's not a matter of fashion, it's a practical reality. No sane business wants to be the who defends the GPL in court. It'll be expensive and messy, and if the result goes against GNU/GPL "accepted wisdom", it will be a PR nightmare.

Nonsense. The GPL is rock solid.
You know how you can tell? It survived the heyday of Microsoft's monopoly without a court challenge.
If Microsoft was afraid to tangle with the GPL at the height of its power, you better believe smaller fish will have an even harder time of it.

No sane business wants to find out what the term "punitive damages" means when trying to violate the GPL for commercial gain.

Re:Chronos, and Apache License thoughts (-1)

Anonymous Coward | about a year and a half ago | (#43210885)

what makes the GPL any different than a porn websites TOS ?

im betting theres more porn sites than GPL projects..

Re:Chronos, and Apache License thoughts (0)

Anonymous Coward | about a year and a half ago | (#43212645)

i'm betting theres more porn sites than GPL projects..

Well, I set out to prove you wrong but I lost count and need a nap and a sandwich.

Re:Chronos, and Apache License thoughts (0)

Anonymous Coward | about a year and a half ago | (#43213311)

The GPL only adds rights you did not have before. If the GPL is not valid... Guess what? You're breaching copyright.

Re:Chronos, and Apache License thoughts (1)

MikeBabcock (65886) | about a year and a half ago | (#43216663)

GPL is not a TOS -- if you reject the GPL but redistribute anyway, you've violated your federal Copyright laws which are often very nasty.

Copyright is the GPL's enforcement as you have none of the rights the GPL gives you without agreeing to it.

Re:Chronos, and Apache License thoughts (0)

Anonymous Coward | about a year and a half ago | (#43210925)

Rock Solid eh? Find me a developer (not a lawyer), who can read through it and understand it in full. Developers generally just select it because they heard good things about it, or have read the summary. And sadly, many of the loud Linux defenders, whilst complaining about the EULA, seem to be ignorant of the difficulty for a standard developer to understand every condition of the license.

Honestly, no developer should be using a license which they cannot EASILY understand themselves. And if they select a difficult-to-read license which is extremely long and difficult to understand , they shouldn't be surprised when developers violate the license, especially since many of these licenses are only available in English (and are difficult to translate).

Re:Chronos, and Apache License thoughts (0)

Anonymous Coward | about a year and a half ago | (#43210963)

Honestly, no developer should ...

So you're saying: Because a developer can't understand legal code, they shouldn't use the license.
Do you also advocate that lawyers who can't understand source code shouldn't use software?

Re:Chronos, and Apache License thoughts (2)

Anne Thwacks (531696) | about a year and a half ago | (#43211413)

Do you also advocate that lawyers who can't understand source code shouldn't use software?

Well, even If he doesn't, I do!

Re:Chronos, and Apache License thoughts (1)

Aaden42 (198257) | about a year and a half ago | (#43213073)

No, but a lawyer should consult with a software engineer before setting out to design software, rather like a software engineer should consult with a lawyer before setting out to apply legal constraints to his creation.

Re:Chronos, and Apache License thoughts (2, Interesting)

Anonymous Coward | about a year and a half ago | (#43211265)

I believe I understand the GPL, all I had to do is to read it. People try to mystify you as "you have to be a lawyer" to read a license if its more than 2 lines. Well nope. The GPL is very clear and uses simple words. It has nothing to do with the gibberish from EULAs, and I guess, that's on purpose.

Re:Chronos, and Apache License thoughts (1)

LordLimecat (1103839) | about a year and a half ago | (#43213697)

You have to be a lawyer to know what the ramifications are, whether it holds up in court, and whether it means what you think it means.

Common sense and the law only overlap some of the time.

Re:Chronos, and Apache License thoughts (2, Insightful)

Anonymous Coward | about a year and a half ago | (#43211309)

What are you talking about? The GPL is both remarkable and robust because it is written in plain english not in twisted legalese. It is easily understandable by developers who take the time to sit down and read it. I encourage you to do that now, maybe you'll change your tune.

Re:Chronos, and Apache License thoughts (0)

geminidomino (614729) | about a year and a half ago | (#43211815)

Did they ever straighten out that whole linking ambiguity?

Re:Chronos, and Apache License thoughts (1)

geminidomino (614729) | about a year and a half ago | (#43223183)

Since my genuine question wasn't answered, but was modded down, (doubtless by some Stallmanite zealot) I'm just gonna go ahead and assume that's a "No."

Not exactly a great example of the GPL's "robustness" there.

Re:Chronos, and Apache License thoughts (1)

MikeBabcock (65886) | about a year and a half ago | (#43216679)

I've met quite a few developers who seem incapable of understanding literate text, so I can't say your assertion is necessary.

That said, the GPL is quite easily readable. For people who don't understand some of its implications, there's lots of online examples to Google just like learning how to program in the first place.

Re:Chronos, and Apache License thoughts (1)

fustakrakich (1673220) | about a year and a half ago | (#43211165)

You know how you can tell? It survived the heyday of Microsoft's monopoly without a court challenge.

That's kinda like saying that I can smoke weed legally because I haven't been busted yet.

Re:Chronos, and Apache License thoughts (0)

Anonymous Coward | about a year and a half ago | (#43211513)

No, it's like saying you can smoke weed legally because you haven't been busted yet, DESPITE the fact that you're always smoking out in the open, and your neighbour, "Mister Someone" would really, really like to see you busted. You know this, because he's been saying, "I'm gonna f*ing KILL (insert your name here)!!!" while throwing chairs across the room.

But some way, somehow, after talking to his lawyer, your neighbour decides not to bust you, but to keep throwing chairs and doing the monkey dance instead.
Why do you think that is?

Re:Chronos, and Apache License thoughts (1)

gringer (252588) | about a year and a half ago | (#43211613)

That's kinda like saying that I can smoke weed legally because I haven't been busted yet.

More like saying you can smoke weed without legal worries because the US government hasn't busted anyone yet.

Re:Chronos, and Apache License thoughts (3, Insightful)

Anonymous Coward | about a year and a half ago | (#43210979)

Stop spouting nonsense. Using GPL software such as a cron scheduler would in no way result in a company going to court.

Re:Chronos, and Apache License thoughts (1)

93 Escort Wagon (326346) | about a year and a half ago | (#43211135)

True. For example: While Apple developed launched and uses it pretty exclusively, they haven't removed cron from OS X. And if you put something in the cron tab, it'll run as expected.

Re:Chronos, and Apache License thoughts (1)

93 Escort Wagon (326346) | about a year and a half ago | (#43211145)

s/launched/launchd/

I love autocorrect... except when I don't.

Re:Chronos, and Apache License thoughts (1)

Aaden42 (198257) | about a year and a half ago | (#43213129)

While Apple developed launchd and uses it pretty exclusively, they haven't removed cron from OS X.

That's rather moot as Apple's cron isn't covered by GPL. They use vixie cron which is BSD licensed.

Re:Chronos, and Apache License thoughts (2)

ciurana (2603) | about a year and a half ago | (#43211141)

Thanks all for your comments.

I'm not arguing for or against the *GPL licenses myself. All I'm saying is that I've experienced enough funding or acquisition due diligence processes to have heard from the acquiring/funding party's counsel that *GPL code must either be replaced with a viable alternative, or that the deal might be called off. Other people have had other experiences, and of course there are companies (e.g. Percona, Red Hat) who are doing well with it.

The ventures in which I've been involved (or the adoption projects that I've supervised, like the official use of open source software at the largest company in the world) have always had a desire to avoid the *GPL as a common denominator. It's accepted only if there is no alternative for the business, or if there is enough education among the developers that the code should not be modified and redistributed in any form, and so on and so forth. You all know how that works.

If the *GPL works for your business model, more power to you. If you find investors/acquires willing to buy your *GPL-reliant product or service -- great! I'd love to hear about that experience too.

Cheers!

pr3d

Re:Chronos, and Apache License thoughts (1)

ciurana (2603) | about a year and a half ago | (#43211179)

By the way -- I don't think legal/business concerns are about the solidity of the license. The concerns are about the aspects that could be hostile to business and investment.

Remember that not everyone wants to make their bacon by offering consulting or other professional services.

Some people want to build and offer finished, successful products that some enterprising licensor may feel are worth pursuing in court over some obscure clause, very much like patent trolls and other IP holders of dubious value go about their business. Avoiding such trolls is a fiduciary obligation of the company management. One way to avoid them is to re-release derivative works, forks, or patches. Another is to avoid software based on such licenses. It's up to each situation and company mission to decide which alternative to pursue.

Deciding whether the company policy regarding a given license is in line one one's beliefs and whether to work with, or for, such a company, are up to each person in terms of his or her professional goals.

Cheers!

pr3d

Re:Chronos, and Apache License thoughts (0)

Anonymous Coward | about a year and a half ago | (#43211259)

*You do your customers a disservice in not explaining to them that their fears about the GPL are in large part unfounded. They are paying you to avoid wasting money and you are failing in filling that role by only telling them what you expect to hear. If the company is unwilling to train their junior devs about what derived work means, that's their own grave.

(* all case to case dependent of course)

I know this sounds pretty harsh, but taking money to propogate what you know is stupidity is a pretty hollow way to make a living and live a life.

Re:Chronos, and Apache License thoughts (1, Informative)

ciurana (2603) | about a year and a half ago | (#43211389)

Like you said, it's case-to-case. And until *GPL is contested in court we won't know for sure.

At Very Large Retailer I engaged Bruce Perens and his team (circa 2006) and we went through education. It paid off. Irrational fear of the GPL was squelched.

At other deals, especially startups, i try to advise them to find the best tech first, then worry about the license, and whenever possible to just avoid *GPL in their products to preempt potential issues. They will have a full plate if they end up in a funding situation. It's my fiduciary obligation to smooth due diligence for them or to protect the investor's interests. If using Apache licensed components reduces friction and increases their chances of having a successful partnership, I would be in breach for that endeavor if I didn't advise them to go with whatever works best for that situation.

At a personal level, I prefer less restrictive licenses than *GPL. I don't know that using or avoiding it is stupidity, as you suggested in your reply. It's just a license, and use or avoidance are specific to each situation. No license is inherently good or bad. As an engineer I rather solve the problem than engage in a phylosophical fight. If *GPL works for some case, then I will suggest to use it.

Cheers!

pr3d4t0r

Re:Chronos, and Apache License thoughts (2)

hweimer (709734) | about a year and a half ago | (#43211539)

And until *GPL is contested in court we won't know for sure.

Come on, this is ridiculous. The GPL has been found perfectly enforcable [wikipedia.org] in many cases in many jurisdictions, with some eventually going to courts. The reason that most cases are settled out of court comes from the fact that defending a GPL violation is such a hopeless endeavor in most situations.

Re:Chronos, and Apache License thoughts (1)

hweimer (709734) | about a year and a half ago | (#43211517)

I'm not arguing for or against the *GPL licenses myself. All I'm saying is that I've experienced enough funding or acquisition due diligence processes to have heard from the acquiring/funding party's counsel that *GPL code must either be replaced with a viable alternative, or that the deal might be called off.

While I understand that this can happen, it effectively means you are advocating against using the GPL not based on the actual content of the license, but because of the (quite likely irrational) behavior of a third party.

Re:Chronos, and Apache License thoughts (1)

ciurana (2603) | about a year and a half ago | (#43215523)

While I understand that this can happen, it effectively means you are advocating against using the GPL not based on the actual content of the license, but because of the (quite likely irrational) behavior of a third party.

If it makes business sense to use *GPL I'll be the first one to advocate it. If there's no reason for it, and an Apache licensed component is available, I'll advocate that. It all depends on the business model and whom I'm advising. If I advise against using code under any particular license is precisely because the license content could have an adverse effect on the business.

I've licensed my own code under GPLv2 when it made sense, under Apache at times, and under BSD most of the time. If we're talking about ideology and personal preference, BSD is the sweet spot for me; the least restrictions of any license.

Cheers!

pr3d

Re:Chronos, and Apache License thoughts (1)

Antique Geekmeister (740220) | about a year and a half ago | (#43212173)

The GPL has been already effectively tested in court, repeatedly. Unfortunately, intellectual property lawyers are scared of the GPL. I discussed it with one 8 days ago: they consider it dangerously viral. I'm trying to arrange a lunch so we can sit down and go over the details of it, so they can understand why I much prefer to use it and I can give examples of companies who pretend to be open source but drive engineer like me nuts when their commercial versions of their "open source" tools break and we can't get source code and can't have any confidence in its security models because they're commercial code is unpublished.

I've also come to loathe the Apache and BSD licenses because when I do work for other groups, I can't easily re-use the work on other projects because I've _already signed_ intellectual property licenses with a previous company under an Apache license. So my work is effectively no longer open source. Many engineers simply ignore such problems, thinking that the development they did on Apache projects is OK to bring to new projects for other companies, and treating Apache licensed development _as if_ they were GPL.

Re:Chronos, and Apache License thoughts (2)

Dr_Barnowl (709838) | about a year and a half ago | (#43212367)

I can't easily re-use the work on other projects because I've _already signed_ intellectual property licenses with a previous company under an Apache license

Out of interest, can you link to the clause in the APL2 license that causes this difficulty? It's not something I heard before and something I hadn't considered.

Re:Chronos, and Apache License thoughts (1)

tlhIngan (30335) | about a year and a half ago | (#43213669)

The GPL has been already effectively tested in court, repeatedly. Unfortunately, intellectual property lawyers are scared of the GPL. I discussed it with one 8 days ago: they consider it dangerously viral. I'm trying to arrange a lunch so we can sit down and go over the details of it, so they can understand why I much prefer to use it and I can give examples of companies who pretend to be open source but drive engineer like me nuts when their commercial versions of their "open source" tools break and we can't get source code and can't have any confidence in its security models because they're commercial code is unpublished.

The problem isn't the GPL per se. Most were fairly happy with the GPLv2. It's GPLv3 that has them in a tizzy - especially on many levels. First is the sheer number of GPLv3 licenses - especially ones like the AGPL that can be problematic.

Next would be the inherent incompatibility - GPLv2 is completely and utterly incompatible with GPLv3, so you cannot mix GPLv2 and GPLv3 code together (especially since both generally go under "GPL" making it hard to determine). GPLv2+ code CAN be mixed with GPLv3 code (forming a GPLv3 work). But then if you had GPLv2 code mixed with GPLv2+ code, and now add GPLv3 code... the whole thing explodes and you've got a nasty mess.

One company I worked at instituted an open-source review program where opensource code would now have to be approved whether it's for use in product code or just internal tools - to speed things up they had lits of pre-approved OSS code. The review presentations made it very clear that GPLv2 was now only allowed by review, BSD and APL would tend to sail through review, and GPLv3, unless preapproved already or one could make a very convincing case, was verboten and will NEVER be approved (short of CEO approval).

Re:Chronos, and Apache License thoughts (0)

Anonymous Coward | about a year and a half ago | (#43212905)

Seems like Chronos is yet another 'Let's re-invent the wheel' project to create a scheduler'. Projects such as Grid Engine do the same thing, and have been doing it very well for a very long time. There are also API interfaces to Grid Engine such as DRMAA (www.drmaa.org) so you can incorporate it into your applications as well.

But since Grid Engine is not written in Java.....the Java guys have to go and write yet another scheduler....that does less, is less scalable, etc...

Re:Chronos, and Apache License thoughts (1)

Beezlebub33 (1220368) | about a year and a half ago | (#43215003)

Seems like Chronos is yet another 'Let's re-invent the wheel' project to create a scheduler'. Projects such as Grid Engine do the same thing, and have been doing it very well for a very long time. There are also API interfaces to Grid Engine such as DRMAA (www.drmaa.org) so you can incorporate it into your applications as well.

Grid Engine is not longer open source or free. They're charging $500/processor (!) for the latest version and have hidden the previously free versions. You might be able to find the earlier versions in other places on the net, but not through Oracle.

But since Grid Engine is not written in Java.....the Java guys have to go and write yet another scheduler....that does less, is less scalable, etc...

As you mention above Grid Engine supports DRMAA, so it works very well with Java from the controller side. The tasks need to be in scripts though, which is a pain.

"nerd blog" (-1)

Anonymous Coward | about a year and a half ago | (#43211477)

... and it fails to show what the config files look like.

Looking further, "After successful installation a local version of Chronos with a built-in ZK server is started. You will need Maven 3.X, a JDK and build tools to get up and running". [github.com] No thanks.

Dependency on a JRE just to schedule jobs ffs? Only if I really, really have to. On JDK? Not a chance.

But... but... (0)

Anonymous Coward | about a year and a half ago | (#43211605)

isn't that systemd's job?

Freshmeat? (1)

helobugz (2849599) | about a year and a half ago | (#43211749)

Seriously, why isn't this on freshmeat instead of /.? How the fawq is this news??

Re:Freshmeat? (0)

Anonymous Coward | about a year and a half ago | (#43215607)

isn't freshmeat dead? I think it has been for years.

Re:Freshmeat? (1)

MikeBabcock (65886) | about a year and a half ago | (#43216691)

Just renamed to freecode ... but yes, its pretty useless now compared to when it was freshmeat.

Control-M (1)

jchawk (127686) | about a year and a half ago | (#43211911)

I'm curious to see how this stacks up against BMC's control-m product. These schedulers are useful when managing 1000's of machines running interdependent jobs.

gas jobs still lingering around (0)

Anonymous Coward | about a year and a half ago | (#43216851)

bank of america GAS Jobs scheduling tool finally made it to opensoure, while the bank tries to rid of it.

Can you trust the submitter with family name: dick (0)

Anonymous Coward | about a year and a half ago | (#43232161)

Hi Guys,
Sometimes i wonder what makes people to put f* kind of words as their last name:
Victor Hooi can be translated into russianh as Victor D**k (soft version).
So what's wrong with people....?

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>