New section for older FreeBSD in forums.

mfaridi said:
I start this post to help users with the old version of FreeBSD to solve their problems.
Click to expand...
You don't seem to get the point. Anyone running any public service on an old OS that's EOL and unsupported already has a huge problem. This is especially true for a service like plesk that allows remote administrative tasks. The only way to solve this problem is to move to a supported, up-to-date operating system. If this, in the case of plesk, indeed means to install a different brand like e.g. Linux, then that's the only way to go, short of abandoning plesk, of course. Anything else is just asking to be "owned".
 
Zirias said:
You don't seem to get the point. Anyone running any public service on an old OS that's EOL and unsupported already has a huge problem. This is especially true for a service like plesk that allows remote administrative tasks. The only way to solve this problem is to move to a supported, up-to-date operating system. If this, in the case of plesk, indeed means to install a different brand like e.g. Linux, then that's the only way to go, short of abandoning plesk, of course. Anything else is just asking to be "owned".
Click to expand...
You are right but help users with old FreeBSD is important, when we say the end of life, they throw away from FreeBSD world.
 
mfaridi said:
when we say the end of life, they throw away from FreeBSD world.
Click to expand...
I doubt that very much, upgrading isn't hard in general. But anyways, I'd strongly prefer that over seeing a lot of cracked outdated FreeBSD boxes. "Helping" with operating an outdated FreeBSD version would give the very wrong impression that this was somehow a reasonable thing to do.
 
mfaridi said:
You are right but help users with old FreeBSD is important, when we say the end of life, they throw away from FreeBSD world.
Click to expand...
I'm pretty sure most of us understand your problem, but I'm not sure you understand the possible issues that your users face using an incredibly old version of FreeBSD.
Helping them move to something more secure, either Plex on another OS or FreeBSD with another tool is the only responsible thing to do.
That said, I know the pain you'd be going through trying to explain this to them.

Just throwing this in here, ever taken a look at sysutils/froxlor?
It's pretty basic, but might get things done.
 
In the real world, we have many many servers running as production systems and does not upgrade cause of cannot work with new OS and developer cannot match their software or application with the new OS, in some telecommunications company, these are happening.
I know the production system with 20 years old age OS and work like charm if something bad happens for the OS or software, they have a nightmare.
 
So, just asking, would you share the network address of said system? Rhethorical question, of course, so, don't ...
 
mfaridi said:
if something bad happens for the OS or software, they have a nightmare.
Click to expand...
Instead of paying top-dollar for support on that ancient software, they better invest in setting up a redundant cluster to replace that software with a newer version.

I see that too in the industry, non-redundant servers that are very old, no body dares touches it afraid it will break down and bring the entire company to a halt. But instead of keeping those ancient non-redundant servers alive, it's better to pull off the band-aid now and invest in setting up new systems which are redundant and are running the latest versions which are being supported. That is the short pain, believe it or not.
 
Having an old system that supports a software application is fine, if the system is not on the network, or the Internet. Once you put a system on a network or worse, on the Internet, you are asking for trouble.
 
Look, we are all volunteers here. I come here because (a) I get my questions answered, often without even asking them, (b) it is amusing, and (c) I can help some people. The question that needs to be asked: Of the limited amount of time I have available for part (c), how should I invest it?

I have ample experience with running up-to-date software, and with running obsolete software. Just in the FreeBSD realm: right now, my system is absolutely up-to-date (I run freebsd-update and pkg upgrade weekly, or if I feel lazy every two weeks), but for about 5 years I ran an obsolete version (got stuck on 9.X for several years, before making a giant leap to 11.X with a complete reinstall). I get it that making a giant reinstall leap is a lot of work, and that's why I have changed how I run my personal FreeBSD system, because spending 15 minutes per week is much less painful than spending 3 solid days after 5 years. But I think the situation that mfaridi describes (where one is stuck on an old OS version because of a piece of application software that is incompatible with newer versions) is actually very rare; in most cases, one is still better off going forward than becoming stuck with old versions.

The real question I have to ask myself is this: Of the limited time I have available for helping people who ask questions here, how should I invest it? I think it's more sensible to help people who have problems with the current version, because those problems can be solved more efficiently. To begin with, if someone asks a question about version 8.X or 9.X, I can't even read the man pages for that easily, much less look at example output from commands. So for better or for worse, I will choose to help people who can be helped easily and efficiently.

If someone really insists on running an old version, they need to accept that this will cause them to have a higher support burden. And that they will have to pay for that support burden, because finding free volunteer support is hard. We, as a community of free volunteers, should rather explain to them that they are better off upgrading, even if it causes short-term pain.
 
One thing I did see was a RocketRaid SATA card, that the box and instructions said it was compatible with FreeBSD 5, 6 or some slightly older version, and it came with a FreeBSD driver. Unfortunately, drivers for that same manufacturer that came with FreeBSD 7 and newer didn't work with that slightly older card. I couldn't get that older driver that came with the box to work on a newer supported FreeBSD either. It was a long time ago when most motherboards only had PATA connections, and that card is obsolete in most cases by motherboard hardware by now.
 
mfaridi said:
In the real world, we have many many servers running as production systems and does not upgrade cause of cannot work with new OS and developer cannot match their software or application with the new OS, in some telecommunications company, these are happening.
I know the production system with 20 years old age OS and work like charm if something bad happens for the OS or software, they have a nightmare.
Click to expand...

Well, it's time to get rid of that old developer that cannot keep up with current operating systems. Let me say it twice - it's time to get rid of that old developer that cannot keep up with current operating systems. I've worked IT in the "real world" for 30 years, and it's a fail of the SA (systems Administrator) of any business, university, company or otherwise who cannot keep their systems that they're responsible for up to date. There is absolutely no excuse.
 
Datapanic said:
Well, it's time to get rid of that old developer that cannot keep up with current operating systems. Let me say it twice - it's time to get rid of that old developer that cannot keep up with current operating systems. I've worked IT in the "real world" for 30 years, and it's a fail of the SA (systems Administrator) of any business, university, company or otherwise who cannot keep their systems that they're responsible for up to date. There is absolutely no excuse.
Click to expand...

Judge not, that ye be not judged.
 
The fact is FreeBSD support time is too short. If I'm developer, even on Linux, I would choose to develop for Redhat or Ubuntu LTS (except vaporware). It takes time, money and effort to write software. I expect I could earn money or at least keep my products functioning long enough and focus on improve stability and features. I don't want to deal with adapting my software to work with new release and you release very often and usually break thing. Your upgrade tool sucks. Anyone still remember or just forgot, the pain of upgrading to 12.0 once occupied this forum? Lua Loader not boot, do you remember it? If not then I should gone.
 
badbrain said:
The fact is FreeBSD support time is too short. If I'm developer, even on Linux, I would choose to develop for Redhat or Ubuntu LTS (except vaporware). It takes time, money and effort to write software. I expect I could earn money or at least keep my products functioning long enough and focus on improve stability and features. I don't want to deal with adapting my software to work with new release and you release very often and usually break thing. Your upgrade tool sucks. Anyone still remember or just forgot, the pain of upgrading to 12.0 once occupied this forum? Lua Loader not boot, do you remember it? If not then I should gone.
Click to expand...
I would kindly refer you to post #6 in this thread. Also, FreeBSD's kernel API's don't change much between versions which makes writing and maintaining software really easy. I don't have enough experience with Linux API's to do a comparison though.
Regarding the freebsd-update tool - as someone using FreeBSD since version 4.x - I disagree with you based on personal experience.
 
Zirias said:
Right .. Linux clearly is the system maintaining stable APIs. You're trolling, but nice try.
Click to expand...
Read my post again. Linux break things, and FreeBSD, too. But we have LTS Linux, at least assurance. On FreeBSD, we have none.
 
So childish. I give you dislike and you give me back one, too. Make me laugh. Every time you see something opposed to your own idea you immediately accuse it will troll, huh? Keep your cult, I left.
 
I'm considering all arguments but... You have some users, despite of what you whish, that use outdated versions. Many of you think that's evil. However, evil is somehow a part of freedom.

I ain't sure the better answer is to close the eyes, to be so affirmative. Not to mention that the evocated problem could be current.
 
badbrain
Please stop bringing up Ubuntu. I get that you like it, but keep that on their forums and elsewhere. This is a FreeBSD forum.

And off topic: I don't see how a bloated system can benefit a professional one. It doesn't, but it's an OS that is the only or few benchmarks for some.
 
badbrain said:
... I would choose to develop for Redhat or Ubuntu LTS (except vaporware). ...
Click to expand...
I don't know about Ubuntu LTS; I don't even remember the last time I installed Ubuntu.

In the case of Redhat, you are comparing apples and oranges. RedHat's distribution is called RHEL, and is not free. You pay to install it, and you pay for support. You are comparing it to FreeBSD, which is in itself free to download. And in this thread we are discussing the free support that this forum provides; the OP mfaridi is of the opinion that this forum should help support older versions, others disagree. Comparing this free support with paid support is ... problematic (a polite term for "trolling").
 
Emrion said:
... use outdated versions. Many of you think that's evil.
Click to expand...
No, I don't think it's evil. On the contrary, for about 5 years I did just that: because I didn't want to invest the (weekly) time to keep current with updates, I deliberately ran an older and older version of FreeBSD on my server. At the time, I thought that was a sensible decision, and I knew the risks involved. Matter-of-fact, if you dig back a year or two, you'll find a thread I started here asking how to go forward, and after getting advice, I decided to reinstall from scratch, which took me about 3 days full-time work (the bulk of that is not the base OS, but getting all the special-purpose stuff and customization stuff back together, including a lot of cleanup of deferred maintenance). From that experience I learned that doing quick updates all the time is the sensible and efficient way to run FreeBSD. Coincidentally, just yesterday I moved my machine to 11.2-p10 (from -p9). I also happen to have some computers at home that run 30-year old software, and I have no plans of upgrading them (even though VMS is still upgradeable).

Running old versions of FreeBSD is a choice. I no longer think it is a good choice, and I don't want to invest my time into helping people making the same mistake I did. But I'm not going hit them with sticks to stop them; it's their freedom.
 
ralphbsz said:
No, I don't think it's evil. On the contrary, for about 5 years I did just that: because I didn't want to invest the (weekly) time to keep current with updates, I deliberately ran an older and older version of FreeBSD on my server. At the time, I thought that was a sensible decision, and I knew the risks involved.
Click to expand...

Same here. I was dependent on ISDN, the ISDN-support was dropped in some Rel.7 or so, and I didn't dare to put the foreign ports from HPSelasky into my kernel, as I feared difficult problems (kernel hacking is not an easy thing, and I didn't have enough time for that). Actually, when I finally tried it, it went completely smooth.

There certainly are valid reasons to stay with an old version, and I am not to judge these reasons (and I couldn't see who were).
I would rather see the whole thing as fun-based: I would help people with old versions if the concerned matter looks interesting me, just the same as with new versions. And I now have written a deployment scheme that does rollout new versions almost automatically (while pulling in all my local stuff and patches), and it is real fun to watch that work (in fact it is rather audible: the fans go fullspeed). So, again, this is no kind of fetish of "thou must <whatever>", but simply fun-based.
 
You must log in or register to reply here.
Back
Top