Took us a while to get here, but this sort of debate and input was exactly what I was hoping for from day one. Ecstatic to see it here, and I want to single out BrutusMaximus and Dryden for really offering outstanding insight, experience and perception based on their relevant expertise. As I've stated many times before, I am definitely no expert. Given my approach to BP, and how I want and need it to be an organism that drives and defines itself, I'm glad you two in particular, but everyone involved so closely in the process (Mils, osu, cleve (who should really be singled out above as well probably), grad, etc. etc. etc.) are playing such an active role. At the end of the day, I have to be the one to pull the trigger, but I wanted the community to aim the gun, does that make any sense?
Understand as well, that my feeling early on is that we shoot past where we *need* to be, towards where we would like to be. This affords some freedoms and luxuries, more importantly it allows for the inevitable hidden expenses and complications that come with something this involved.
Let's build the machine. I'd like to see you two (and anyone else) hash out something that can collectively be agreed upon. I'd also like us to start narrowing down the best places to purchase from (I know all of this has already started). And then I need you to really start thinking about the level of help I'm going to need to get it configured as BP needs it. I've installed FreeBSD (as an example) before. Tried to walk myself through building a custom kernal and all of that. I was successful, but it was neither clean nor pretty. I like clean, and I like pretty. What I'm getting at here, is given enough time, I'll figure anything out and force myself through a learning process to a functional level. But we want to go beyond that for the configuration of this machine, and I'm hoping one or both of you, or someone else, will be able to step up and play an active role in its creation, as well as its ongoing maintenance. I know this skirts right along the edge of asking you to do what you do for income, except for free. I'm not looking for anyone to be my beck-and-call tech monkey, but be available to do things that need to be done at your leisure and convenience.
Here's the catch with everything. Imagine that we're tossing the machine into a vacuum when it's done. Meaning, it's a major ordeal in order to get someone to physically do anything to the machine. Including something as simple as power cycle it. But it's worth that price, when the reward is a free resource we can always count on -- never subject to personal, political, or other external issues. So talk about putting the machine together with that in mind. Adding RAM later is bound to be a major undertaking, which is fine, but it will be no simple matter. If something craps out, it's wonderful that a company can get a replacement part there the next day. The problem is it may sit there for 3 weeks before I can get someone to actually install it. We have zero support at this NOC. I can call in the cavalry now and again, but as we're neither the core business functioning out of there, nor are we a paying 3rd party client taking advantage of their ridiculous resources -- we are on the bottom rung of the priority ladder. Check that, we're not even on the ladder.
My concern is this fall. To that end, I'm more interested in being prepared for next year, than I am about being prepared for the year after. However, if we have the option of being better prepared for the future, than I'm all for that.
Anyway, I'm rambling now. There are still some outstanding question marks related to how much we have to work with. However, based on numbers I've recently seen tossed around in here, we *are* at a point where we know we could handle at least that. We can start making hard decisions on issues at hand.
In regards to (insert major corporation name here) building our machine and us having a service contract -- vs. us building it by shopping for our own parts, we just have to be cognizant of the support issue. If there are people on the site I can turn to for information and even tangible assistance, then I don't need a support structure from (insert major corporation name here). Having the safety net that someone from (insert major corporation name here) will go to the machine and resolve problems locally is a nice thing (assuming I could get them in the door -- there *are* lots of people working in and around that facility, but they don't get paid to work on my box), but it is by no means an absolute need.
After all, little cinnamon (not my choice) has been sitting there all along, chugging away, without a power cycle in well over a year. No cycle because I have reason to believe it wouldn't come back up, lol, but this is what we're going to ask a machine to do. Happily, the environment is perfect, so if we do a good enough job picking parts, and configuring it, we should never have a problem. But hopefully I'm painting a clear enough picture, in regards to just how much we need to value reliability, redundancy, layered comfort and safety, etc.
STILL babbling, some things never change, eh? Outstanding job in this thread, *never* feel like you need to stop providing input and opinion, I not only want you to do so, but I expect it of you (where 'you' is certainly the whole community, but I'm pointing it at this moment squarely at BMax and Dry). Let's hash the rest of it out, disagree and debate all you like, that's the best way to reach a consensus on the issues.
I will make sure I'm more directly available in the short term to help keep the process rolling, so those who have been using Jo as a point of contact for me, can shift these issues to my address (my first name in the short form at the same dot net) and I will do my best to be as responsive as possible, allowing of course for some downtime and delay related to issues on my end that reach into the extremes, here and there. To this end, I've subscribed to the thread so I will receive an email every time another post is added to it. I still love the bells and whistles here, even if many are silly. :)
Understand as well, that my feeling early on is that we shoot past where we *need* to be, towards where we would like to be. This affords some freedoms and luxuries, more importantly it allows for the inevitable hidden expenses and complications that come with something this involved.
Let's build the machine. I'd like to see you two (and anyone else) hash out something that can collectively be agreed upon. I'd also like us to start narrowing down the best places to purchase from (I know all of this has already started). And then I need you to really start thinking about the level of help I'm going to need to get it configured as BP needs it. I've installed FreeBSD (as an example) before. Tried to walk myself through building a custom kernal and all of that. I was successful, but it was neither clean nor pretty. I like clean, and I like pretty. What I'm getting at here, is given enough time, I'll figure anything out and force myself through a learning process to a functional level. But we want to go beyond that for the configuration of this machine, and I'm hoping one or both of you, or someone else, will be able to step up and play an active role in its creation, as well as its ongoing maintenance. I know this skirts right along the edge of asking you to do what you do for income, except for free. I'm not looking for anyone to be my beck-and-call tech monkey, but be available to do things that need to be done at your leisure and convenience.
Here's the catch with everything. Imagine that we're tossing the machine into a vacuum when it's done. Meaning, it's a major ordeal in order to get someone to physically do anything to the machine. Including something as simple as power cycle it. But it's worth that price, when the reward is a free resource we can always count on -- never subject to personal, political, or other external issues. So talk about putting the machine together with that in mind. Adding RAM later is bound to be a major undertaking, which is fine, but it will be no simple matter. If something craps out, it's wonderful that a company can get a replacement part there the next day. The problem is it may sit there for 3 weeks before I can get someone to actually install it. We have zero support at this NOC. I can call in the cavalry now and again, but as we're neither the core business functioning out of there, nor are we a paying 3rd party client taking advantage of their ridiculous resources -- we are on the bottom rung of the priority ladder. Check that, we're not even on the ladder.
My concern is this fall. To that end, I'm more interested in being prepared for next year, than I am about being prepared for the year after. However, if we have the option of being better prepared for the future, than I'm all for that.
Anyway, I'm rambling now. There are still some outstanding question marks related to how much we have to work with. However, based on numbers I've recently seen tossed around in here, we *are* at a point where we know we could handle at least that. We can start making hard decisions on issues at hand.
In regards to (insert major corporation name here) building our machine and us having a service contract -- vs. us building it by shopping for our own parts, we just have to be cognizant of the support issue. If there are people on the site I can turn to for information and even tangible assistance, then I don't need a support structure from (insert major corporation name here). Having the safety net that someone from (insert major corporation name here) will go to the machine and resolve problems locally is a nice thing (assuming I could get them in the door -- there *are* lots of people working in and around that facility, but they don't get paid to work on my box), but it is by no means an absolute need.
After all, little cinnamon (not my choice) has been sitting there all along, chugging away, without a power cycle in well over a year. No cycle because I have reason to believe it wouldn't come back up, lol, but this is what we're going to ask a machine to do. Happily, the environment is perfect, so if we do a good enough job picking parts, and configuring it, we should never have a problem. But hopefully I'm painting a clear enough picture, in regards to just how much we need to value reliability, redundancy, layered comfort and safety, etc.
STILL babbling, some things never change, eh? Outstanding job in this thread, *never* feel like you need to stop providing input and opinion, I not only want you to do so, but I expect it of you (where 'you' is certainly the whole community, but I'm pointing it at this moment squarely at BMax and Dry). Let's hash the rest of it out, disagree and debate all you like, that's the best way to reach a consensus on the issues.
I will make sure I'm more directly available in the short term to help keep the process rolling, so those who have been using Jo as a point of contact for me, can shift these issues to my address (my first name in the short form at the same dot net) and I will do my best to be as responsive as possible, allowing of course for some downtime and delay related to issues on my end that reach into the extremes, here and there. To this end, I've subscribed to the thread so I will receive an email every time another post is added to it. I still love the bells and whistles here, even if many are silly. :)
Last edited:
Upvote
0
