No it's not. You're assuming that they "own" the complete, or nearly complete, market between the two of them. There's nothing wrong with sharing technology and having an agreement to compete with other companies in certain areas. Now, if between the two of them they "owned" 90% of the market then the situation would be a little more complicated...
You'd better not mention your opinion on the linux-kernel mailing list. If you think/. comments are bad! (But I agree with the vocal responses on l-k and would likely join in the process of "educating" you how stupid an idea putting a database in the kernel of an OS is for a general purpose OS).
Now if you're talking about a specialized "database" appliance box then that's different. But for general purpose OS's integrating a database is just plain stupid.
As a matter of fact, I'm using ITO 5.0 (NNM 6.0) and I was speaking specifically about ITO's application monitoring capabilities (or lack of).
Do you have all the latest patches?
We are using ITO on a correctly sized HP/UX server, thank you.
Good!
Now, I'd like to know people using IP unnumbered interfaces on their routers (which Cisco says is the proper way of configuring WAN links) and still don't poll the routing table.
Try editing netmon.lrf and adding -R to the 3rd field, ovdelobj netmon.lrf, ovaddobj netmon.lrf, ovstop netmon, ovstart netmon. That will tell netmon not to regularly poll routing tables on routers.
Using unnumbered interfaces for WAN links may be the way cisco recommends configuring their routers, but it is not, IMNSHO, the best method. If someone told you to jump off a bridge would you? Using numbered interfaces gives you the ability to ping the other side whereas you don't have that ability using unnumbered interfaces. You can check each level of the stack to see where the problem lies, instead of just having everything "Up" or "Down." I don't recommend using unnumbered interfaces to anyone. (Just like I recommend against using zero subnet networks, even though all router vendors I know of support it).
Why would you want to poll the routing table with unnumbered interfaces? To see if they are up or down? If that's the case, the default for NNM is to SNMP poll unnumbered interfaces. In ITO that default may be different, but I doubt it. If it is, then you can add a -k nonIPStatusPolls=true to the netmon.lrf file. This will give you status info on your unnumbered interfaces. If you don't see connections between your unnumbered interfaces, you could add -k discoverLevel2Nets=true to netmon.lrf. Again, this is the default, so someone must have been mucking around, or ITO a lot different than the base NNM, if you have to do this to get acceptable results.
I again encourage you to join the OVForum. You don't need to be a member (which requires a membership fee) in order to join their mailing list....
Well, there's options to netmon to turn off pulling the routing table each status poll. This is pretty well know for competent HP OpenView users.
As far as applications monitoring, that's not what NNM is for. Take a look at ITO if you want to do systems management. HP OpenView Network Node Manager is for network monitoring, not applications monitoring. FYI, ITO includes NNM as it's "base" platform.
You're not using the Windows NT version are you? That would be your first mistake. The daemons are not much of a problem if you have a properly sized Unix box. Did you follow the suggestions in the sizing white paper about how much RAM/swap/disk you should have depending on the number of nodes/ovw sessions/etc? Are you using an external database or the bundled one? For large installations you probably want to use Oracle as the database.
Sounds like you are just a little inexperienced and are complaining because you can't figure it out. As a first-step help. I'd strongly recommend joining the HP OpenView Forum and definately their mailing list. The problems you describe are "newbie" problems that get covered quite often on the list (such as polling the router tables in ISP routers). I think you'll get up to speed rather quickly if you participate in the discussions on the mailing list.
Anyone remember when the scientist had a fear of igniting the atmosphere when they were testing the H-bomb? I always thought that one was funny also. Or was that some kind of Cobalt bomb or something? Geesh, I'm loosing it.
But, it's kinda the same with all the companies that Microsoft has bought out to "stifle competition." A certain amount of blame has to be put on the stupid little companies that make the agreements with Microsoft and then find themselves in a bad position. It's almost like some of them have absolutely no foresight, and a terrible memory. What's amazing to me is that there are STILL companies that dont' realize that Microsoft will take advantage of them given the chance.
Apple may be similar to a certain extent, but you can't totally obsolve Sorenson of their involvement in the situation... That's the point I was trying to make. We can still beat Apple over the head and petition them to create a Linux client, but let's beat Sorenson over the head too!
Computing one key involves much more than one instruction. I doubt the e2k could do as many instructions as necessary for even one key per cycle, let alone 3 keys per cycle. What are we talking about, hundreds or thousands of instructions per key?
Sorenson is saying this because they want to avoid any heat for the decisions they have made. If they entered into an agreement with Apple (signed, legal, or not) that basically says that Sorenson will not release the specs for any platform that does not have QuickTime ported to it then that's their fault. They need to take responsibility for their own decisions. It's their fault for trying to "corner the market" into their proprietary protocol by making such a stupid deal with a computer manufacturer to promote their technology.
Even if it is in a legal document signed by both Apple and Sorenson, I'd have a hard time believing that Apple could do anything if Sorenson did give out the specs. All it could possibly do is make the agreement between Apple and Sorenson null and void. It's not like Apple could sue Sorenson over the release. That would be political suicide anyway. They'd never live it down, and all their talk about opensource would be for null.
So, it's Sorenson's fault for getting into a stupid, idiotic agreement with Apple and not having the balls to break the agreement. And they have the nerve to whine that "Apple won't allow [us] to license it to others!" Oh, bullshit.
Apple doesn't have control, the maker does. How could Apple "not let them?" Are you referring to the agreement that Sorenson and Apple have in their "you scratch my back and I'll scratch yours" policy explained here? If so, then you have to blame Sorenson just as much as Apple. As suggested, write to Charles and tell him why Apple should port to Linux.
As I replied jsut 30 secs ago, that's an old version that doesn't contain the "new" technologies. It hasn't been kept up to date since Microsoft bought out the company that produced it.
Yes, but the Microsoft client is old. It doesn't have all the capabilities of the new clients, and isn't being kept up to date with the main development tree. It was developed (ported, whatever) by the company that Microsoft bought out to get the code. Microsoft dropped support for it as fast as they could. That's why you don't see download links to it prominently posted all over the place. It's old technology that isn't as useful anymore.
I don't think it will be too long before a Linux client pops up...
Oh, God. How may times do we have to go through this? Apple CAN'T release the specs. They lease the specs from another company. It's not their specs to release. Complain to someone else.
I remember seeing a beta of mediaplayer for linux on the ms site last year...
I'll see if i can find the link
Yea, I remember seeing that also. But that still doesn't explain why you "don't think Quicktime is good for linux." Why? Do you think Microsoft's technology is better?
That's interesting -- didn't know that. Doesn't seem like a particularly good deal from MS' perspective anymore. Yes, they could be positioned to make quite a bit of money if Apple continues on it's "comeback." But what happens if that comeback involves taking business away from Microsoft? At levels greater than what MS would make in their stock holdings? MS would end up "loosing" money on the deal, even though they would be making money "on the books."
The interesting question is what happens if they succeed in leveling the marketplace, and making microsoft not so much of a threat. Will the benevolence to Linux continue? Or will Linux be seen as a threat and become the next target?
I don't think Linux will ever be thought of as a "threat." There is no one company that anyone can "target" in the Linux community. Anyone can use Linux and make their own distribution. SGI, IBM, Oracle, and Sun can all make their own distributions and fight against each other for the best features/support. That's the beauty of the community, it fosters higher levels of competition rather than promoting proprietary solutions to "lock in" you customer base. Sooner or later, someone will come out with a product that outweighs the benefits of yours and the pain in totally replacing a proprietary solution. Your customers are never "locked in" forever....
And I don't think it will take *years* to play out...
Doesn't MS own somewhere around 10% of Apple? Do they have representation on the board? Don't they have a say in overall company direction? If apple is taking actions that severely effect the finances of own of their primary stock holders, don't you think something would be done.
Oh, but you have to think of the "politically correct" way of handling this also. I can just hear MS whining now, "But Steve, all that high-quality software you are releasing as OSS is taking away our business that we've spent billions in developing!" I don't think that will every happen...
To support real time streams, Darwin includes a real time kernel, something not available in Windows NT, Linux, or anywhere else.
Well, I think the QNX folks have realtime kernels, and I believe there is a Linux RT or RT Linux project or something that brings RT capabilities to Linux. I also believe there is a realtime priority on NT, although I don't know how realtime it really is. Anyone have more info?
Oddly enough, were there ever a nuclear war, much of the US defense would be decimated by EMPs from the first attack wave while Russian tube controlled systems would be far far less vulnerable to EMPs and thus would continue to function effectively. Good thing this was never put to the test, though.:/
Slashdot Mirror
No it's not. You're assuming that they "own" the complete, or nearly complete, market between the two of them. There's nothing wrong with sharing technology and having an agreement to compete with other companies in certain areas. Now, if between the two of them they "owned" 90% of the market then the situation would be a little more complicated...
Can you say FreshMeat?
You'd better not mention your opinion on the linux-kernel mailing list. If you think /. comments are bad! (But I agree with the vocal responses on l-k and would likely join in the process of "educating" you how stupid an idea putting a database in the kernel of an OS is for a general purpose OS).
Now if you're talking about a specialized "database" appliance box then that's different. But for general purpose OS's integrating a database is just plain stupid.
As a matter of fact, I'm using ITO 5.0 (NNM 6.0) and I was speaking specifically about ITO's application monitoring
capabilities (or lack of).
Do you have all the latest patches?
We are using ITO on a correctly sized HP/UX server, thank you.
Good!
Now, I'd like to know people using IP unnumbered interfaces on their routers (which Cisco says is the proper way of
configuring WAN links) and still don't poll the routing table.
Try editing netmon.lrf and adding -R to the 3rd field, ovdelobj netmon.lrf, ovaddobj netmon.lrf, ovstop netmon, ovstart netmon. That will tell netmon not to regularly poll routing tables on routers.
Using unnumbered interfaces for WAN links may be the way cisco recommends configuring their routers, but it is not, IMNSHO, the best method. If someone told you to jump off a bridge would you? Using numbered interfaces gives you the ability to ping the other side whereas you don't have that ability using unnumbered interfaces. You can check each level of the stack to see where the problem lies, instead of just having everything "Up" or "Down." I don't recommend using unnumbered interfaces to anyone. (Just like I recommend against using zero subnet networks, even though all router vendors I know of support it).
Why would you want to poll the routing table with unnumbered interfaces? To see if they are up or down? If that's the case, the default for NNM is to SNMP poll unnumbered interfaces. In ITO that default may be different, but I doubt it. If it is, then you can add a -k nonIPStatusPolls=true to the netmon.lrf file. This will give you status info on your unnumbered interfaces. If you don't see connections between your unnumbered interfaces, you could add -k discoverLevel2Nets=true to netmon.lrf. Again, this is the default, so someone must have been mucking around, or ITO a lot different than the base NNM, if you have to do this to get acceptable results.
I again encourage you to join the OVForum. You don't need to be a member (which requires a membership fee) in order to join their mailing list....
HTH
Well, there's options to netmon to turn off pulling the routing table each status poll. This is pretty well know for competent HP OpenView users.
As far as applications monitoring, that's not what NNM is for. Take a look at ITO if you want to do systems management. HP OpenView Network Node Manager is for network monitoring, not applications monitoring. FYI, ITO includes NNM as it's "base" platform.
You're not using the Windows NT version are you? That would be your first mistake. The daemons are not much of a problem if you have a properly sized Unix box. Did you follow the suggestions in the sizing white paper about how much RAM/swap/disk you should have depending on the number of nodes/ovw sessions/etc? Are you using an external database or the bundled one? For large installations you probably want to use Oracle as the database.
Sounds like you are just a little inexperienced and are complaining because you can't figure it out. As a first-step help. I'd strongly recommend joining the HP OpenView Forum and definately their mailing list. The problems you describe are "newbie" problems that get covered quite often on the list (such as polling the router tables in ISP routers). I think you'll get up to speed rather quickly if you participate in the discussions on the mailing list.
HTH
Anyone remember when the scientist had a fear of igniting the atmosphere when they were testing the H-bomb? I always thought that one was funny also. Or was that some kind of Cobalt bomb or something? Geesh, I'm loosing it.
Exactly! But if you're going to start placing blame, place blame at the feet of both Apple and Sorenson and not just Apple.
Hopefully, that PHB will be looking for another job soon.
I'm not trying to defend Microsoft here
But, it's kinda the same with all the companies that Microsoft has bought out to "stifle competition." A certain amount of blame has to be put on the stupid little companies that make the agreements with Microsoft and then find themselves in a bad position. It's almost like some of them have absolutely no foresight, and a terrible memory. What's amazing to me is that there are STILL companies that dont' realize that Microsoft will take advantage of them given the chance.
Apple may be similar to a certain extent, but you can't totally obsolve Sorenson of their involvement in the situation... That's the point I was trying to make. We can still beat Apple over the head and petition them to create a Linux client, but let's beat Sorenson over the head too!
Computing one key involves much more than one instruction. I doubt the e2k could do as many instructions as necessary for even one key per cycle, let alone 3 keys per cycle. What are we talking about, hundreds or thousands of instructions per key?
Read this for my response to your failure to understand that it's not Apple's fault that Sorenson got into a bad deal with them.
Stop blaming the devil for convincing Eve to take a bite of the Apple and start blaming Eve...
Sorenson is saying this because they want to avoid any heat for the decisions they have made. If they entered into an agreement with Apple (signed, legal, or not) that basically says that Sorenson will not release the specs for any platform that does not have QuickTime ported to it then that's their fault. They need to take responsibility for their own decisions. It's their fault for trying to "corner the market" into their proprietary protocol by making such a stupid deal with a computer manufacturer to promote their technology.
Even if it is in a legal document signed by both Apple and Sorenson, I'd have a hard time believing that Apple could do anything if Sorenson did give out the specs. All it could possibly do is make the agreement between Apple and Sorenson null and void. It's not like Apple could sue Sorenson over the release. That would be political suicide anyway. They'd never live it down, and all their talk about opensource would be for null.
So, it's Sorenson's fault for getting into a stupid, idiotic agreement with Apple and not having the balls to break the agreement. And they have the nerve to whine that "Apple won't allow [us] to license it to others!" Oh, bullshit.
so?
That's because the MS lawyers are a bunch of idiots. Wait.... Maybe that's their basis for appeal, incompetant representation...
Apple doesn't have control, the maker does. How could Apple "not let them?" Are you referring to the agreement that Sorenson and Apple have in their "you scratch my back and I'll scratch yours" policy explained here? If so, then you have to blame Sorenson just as much as Apple. As suggested, write to Charles and tell him why Apple should port to Linux.
As I replied jsut 30 secs ago, that's an old version that doesn't contain the "new" technologies. It hasn't been kept up to date since Microsoft bought out the company that produced it.
Yes, but the Microsoft client is old. It doesn't have all the capabilities of the new clients, and isn't being kept up to date with the main development tree. It was developed (ported, whatever) by the company that Microsoft bought out to get the code. Microsoft dropped support for it as fast as they could. That's why you don't see download links to it prominently posted all over the place. It's old technology that isn't as useful anymore.
I don't think it will be too long before a Linux client pops up...
Oh, God. How may times do we have to go through this? Apple CAN'T release the specs. They lease the specs from another company. It's not their specs to release. Complain to someone else.
I remember seeing a beta of mediaplayer for linux on the ms site last year...
I'll see if i can find the link
Yea, I remember seeing that also. But that still doesn't explain why you "don't think Quicktime is good for linux." Why? Do you think Microsoft's technology is better?
After all, its most likely for MS to release a linux port of media player than for apple to realease a quicktime client.
Why do you say this?
See this comment. Microsoft may end up loosing on their investment even though Apple stock is doing better.
That's interesting -- didn't know that. Doesn't seem like a particularly good deal from MS' perspective anymore. Yes, they could be positioned to make quite a bit of money if Apple continues on it's "comeback." But what happens if that comeback involves taking business away from Microsoft? At levels greater than what MS would make in their stock holdings? MS would end up "loosing" money on the deal, even though they would be making money "on the books."
The interesting question is what happens if they
succeed in leveling the marketplace, and making
microsoft not so much of a threat. Will the
benevolence to Linux continue? Or will Linux
be seen as a threat and become the next target?
I don't think Linux will ever be thought of as a "threat." There is no one company that anyone can "target" in the Linux community. Anyone can use Linux and make their own distribution. SGI, IBM, Oracle, and Sun can all make their own distributions and fight against each other for the best features/support. That's the beauty of the community, it fosters higher levels of competition rather than promoting proprietary solutions to "lock in" you customer base. Sooner or later, someone will come out with a product that outweighs the benefits of yours and the pain in totally replacing a proprietary solution. Your customers are never "locked in" forever....
And I don't think it will take *years* to play out...
Doesn't MS own somewhere around 10% of Apple? Do they have representation on the board? Don't they have a say in overall company direction? If apple is taking actions that severely effect the finances of own of their primary stock holders, don't you think something would be done.
Oh, but you have to think of the "politically correct" way of handling this also. I can just hear MS whining now, "But Steve, all that high-quality software you are releasing as OSS is taking away our business that we've spent billions in developing!" I don't think that will every happen...
To support real time streams, Darwin includes a real time kernel,
something not available in Windows NT, Linux, or anywhere else.
Well, I think the QNX folks have realtime kernels, and I believe there is a Linux RT or RT Linux project or something that brings RT capabilities to Linux. I also believe there is a realtime priority on NT, although I don't know how realtime it really is. Anyone have more info?
Oddly enough, were there ever a nuclear war, much of the US defense would be decimated :/
by EMPs from the first attack wave while Russian tube controlled systems would be far far less vulnerable to EMPs and thus would continue to
function effectively. Good thing this was never put to the test, though.
All our missiles are hardened.