Tell me how to "Work around" an issue where something is so broken as to make it unusable in a production environment besides scrapping it and or rewriting it. This is not always a case of "Our code does not behave as expected" but could be as simple as "The firmware for device X is expecting X and receives Y. This causes it to fail and the connection drops while the server is still waiting"
This could even be for something as simple as "Dell XYZ works this way, but now we have a contract with HP for servers and the kernel is behaving differently. The hardware vendor blames the software vendor and vice versa. Get a support contract or at least go with something that is supportable. I would not even blame people who go out and get 10 licenses for RHEL and then 30 copies of CentOS for 40 machines.
it is not a.deb vs.rpm debate. The debian repositories are known for having a breadth of different packages. If you need one of the more obscure ones, having it built already for your OS and architecture would work for places that won't allow you to compile your own or get it from an outside third party.
Even if Canonical DID do this, they would still be outclassed by SLED. Novell, love them or hate them, has some of the best interoperability out of the box with a Windows environment. I found Ubuntu to be "nice" but fragile. Many of the reasons for its use have been eradicated, and many still exist. The debian package tree is great, but not enough of a reason for me, as I can compile my own source or even make source rpms and then package rpms. YaST is a great tool and the SUSE repos are great for my needs.
Also, while I have gotten some good help on the Ubuntu forums, it is SERIOUSLY a case of the blind leading the more blind. If they really wanted to fix their problems, they should give weight to the different posters on knowledge. Having somebody give out bad configuration that they kludged to work, albeit with major repercussions and possible security holes is NOT a standard by which to judge support
I agree wholeheartedly. If I were to deploy Linux on a corporate scale project, and use Windows admins, I would be fired for incompetence. The admins for the role should be adequately trained on the systems they manage. Either send your Windows guys to Linux training, or replace them with what you actually need.
Quoting bullshit figures is not an alternative to having an argument. 99.999% is a made up number. When the kernel that ships with the product has flaws with the configuration that is being used, for whatever reason, the vendor should try to do their best to fix the situation.
When I was administering a Novell/SUSE network, and we had issues where SAMBA would drop kerberos tickets in our environment, Novell provided us with a custom package for SAMBA to fix the errors.
In another situation on RHEL, Red Hat provided patches for OUR company to fix issues we had with Red Hat Cluster.
Just because you have never hit on interoperability or configuration issues that make and break business does not mean it is not important. Just because you think having an instance of Apache running, without load balancing application routers doesn't mean that is how the enterprise world works. There are a LOT of Oracle App and DB servers on Linux. RAC is very popular as is Oracle 9i and 10g database. Being ignorant does not make you right.
Forums and help communities are great for testing beds or for non critical-path errors. If I need configuration help but don't want to waste time on a phone call for something trivial, I can post to a forum for help, and do get it. However, what would you tell your boss when your main DB server is tossing out errors and refusing connections, but you don't have paid support?
"Hey, don't worry. I posted at 9AM. In a few hours, somebody will respond with something that may fix the problem" doesn't seem to cut it in that scenario
The reason they use it for smaller companies is that they are probably NOT paying for support and don't call for things like kernel fixes or package fixes. What kind of support does Ubuntu have for tools when not even all versions of RHEL or SLES, let alone Oracle Linux are supported? Where is the OMSA package for Ubuntu?
The name helps sell PHBs, but the support from either RH or Novell is far better. I am sure Canonical can do well, but will they put boots on the ground in enough time to support outages?
What is the model for cloning machines, deploying machines and such?
What is the structure for connecting to various directories?
What do they have to offer, besides the.deb repositories and less long term support, than Novell/SUSE and Red Hat or Oracle cannot do now?
They are late to the party, and while I am glad for the strides they have made, Novell and Red Hat can eat them for lunch with other tie ins with their product line.
This is news?? HR has always been "A part of the company."
This is like when your parents tell you to tell the truth and you won't be punished. You get punished anyhow. HR is the business' way of covering its own ass.
I have commented on them a lot, namely the default block size for reads as well as the ARC buffer size issues that need kernel tuning to fix, but SHOULD be fixed, and after only 5 updates to Solaris 10.....
btrfs should be an option on Linux, for those who care to go that route
They got rid of VPR? I was using a very old OV on Solaris 8 or 9 (280R). I never got the love for SPARC but I am a Linux and HPUX guy, myself. I wish I knew Solaris better, but it always felt so different just for the sake of being different.
The other thing I found with OV was that it never gets set up properly. Often alerts are suppressed that are not meant to, things get missed, or it gives you so much to wade through you may as well miss alerts. The issue is not just the software, but the methodology. The people who actually NEED the alerts have to get them from people who are often paid screen watching monkeys who have little to no idea what any of it means, and are paid poorly and treated poorly and then don't do a good job.
My company is now paying HP to come back and tell us how to improve our process, meanwhile if they actually listened to the employees and contractors, they would know exactly what we need to do. Oh well, guess we need to go fill out a code for that (Man, do I hate having to do stupid arbitrary paperwork for the most mundane crap).
Having been in Openview shops exclusively, I can see how powerful, and yet frickin' misguided it is. Openview is used, not because of how powerful or elegant it is (it is one, but not the other) but because "HP makes it, and everybody uses it, so we are taking less risk."
I have never been in an environment that actually used Openview and LIKED it, and the alerts they got. Often times, the operator at the console sees a scroll of gibberish and just recognizes that the reds and oranges are meant to be that way , and ignores them for the most part, EXCEPT when they are meaningless.
I have been called at 3AM because the status of a node changed from NORMAL to NORMAL.......
However, when we lost 90% of our data center to over heating from a total A/C failure......well, they just assumed an outage that big was a scheduled one they were not made aware of and ignored it, due to it filling their screen up......
All the NMS software in the world can't fix lazy, stupid or overwhelmed.
I have not used whatsup for years and even then, just a demo. I liked hyperic for the free version and in some environments, especially the OpenView and Tivoli ones, the $800 per 2 CPU would be about an 70% discount.
There's a free version of the software. You lose some things like forming groups, etc but it is much better than "whatsup" as it can go on a per service/script scan. The Free version also loses out on the "remediation," as such that if you have a condition like "If apache is not up, run this cleanup and then start apache, and notify this group" Also, Hyperic can give you discounts and even charges half for non production environments.
The non enterprise can be used in a 30 server environment, but you DO give up a lot of functionality and have to re-invent things for it to work. I just wish they open sourced the lot of it, but unlikely
You could always try to see if you could be an IT recruiter. That way, you could be on of the people that vets candidates for jobs and get a tidy little commission out of it. The benefit of it is that you can see what jobs are on the market in your field, as well as making sure that incompetent people have a harder time entering it.
Slashdot Mirror
This could even be for something as simple as "Dell XYZ works this way, but now we have a contract with HP for servers and the kernel is behaving differently. The hardware vendor blames the software vendor and vice versa. Get a support contract or at least go with something that is supportable. I would not even blame people who go out and get 10 licenses for RHEL and then 30 copies of CentOS for 40 machines.
it is not a .deb vs .rpm debate. The debian repositories are known for having a breadth of different packages. If you need one of the more obscure ones, having it built already for your OS and architecture would work for places that won't allow you to compile your own or get it from an outside third party.
Also, while I have gotten some good help on the Ubuntu forums, it is SERIOUSLY a case of the blind leading the more blind. If they really wanted to fix their problems, they should give weight to the different posters on knowledge. Having somebody give out bad configuration that they kludged to work, albeit with major repercussions and possible security holes is NOT a standard by which to judge support
I agree wholeheartedly. If I were to deploy Linux on a corporate scale project, and use Windows admins, I would be fired for incompetence. The admins for the role should be adequately trained on the systems they manage. Either send your Windows guys to Linux training, or replace them with what you actually need.
When I was administering a Novell/SUSE network, and we had issues where SAMBA would drop kerberos tickets in our environment, Novell provided us with a custom package for SAMBA to fix the errors.
In another situation on RHEL, Red Hat provided patches for OUR company to fix issues we had with Red Hat Cluster.
Just because you have never hit on interoperability or configuration issues that make and break business does not mean it is not important. Just because you think having an instance of Apache running, without load balancing application routers doesn't mean that is how the enterprise world works. There are a LOT of Oracle App and DB servers on Linux. RAC is very popular as is Oracle 9i and 10g database. Being ignorant does not make you right.
"Hey, don't worry. I posted at 9AM. In a few hours, somebody will respond with something that may fix the problem" doesn't seem to cut it in that scenario
Brown. It is full of brown. What can brown do for you?
The name helps sell PHBs, but the support from either RH or Novell is far better. I am sure Canonical can do well, but will they put boots on the ground in enough time to support outages?
What is the model for cloning machines, deploying machines and such?
What is the structure for connecting to various directories?
They are late to the party, and while I am glad for the strides they have made, Novell and Red Hat can eat them for lunch with other tie ins with their product line.
This is like when your parents tell you to tell the truth and you won't be punished. You get punished anyhow. HR is the business' way of covering its own ass.
btrfs should be an option on Linux, for those who care to go that route
I have to level my camping skill first
The other thing I found with OV was that it never gets set up properly. Often alerts are suppressed that are not meant to, things get missed, or it gives you so much to wade through you may as well miss alerts. The issue is not just the software, but the methodology. The people who actually NEED the alerts have to get them from people who are often paid screen watching monkeys who have little to no idea what any of it means, and are paid poorly and treated poorly and then don't do a good job.
My company is now paying HP to come back and tell us how to improve our process, meanwhile if they actually listened to the employees and contractors, they would know exactly what we need to do. Oh well, guess we need to go fill out a code for that (Man, do I hate having to do stupid arbitrary paperwork for the most mundane crap).
I have never been in an environment that actually used Openview and LIKED it, and the alerts they got. Often times, the operator at the console sees a scroll of gibberish and just recognizes that the reds and oranges are meant to be that way , and ignores them for the most part, EXCEPT when they are meaningless.
I have been called at 3AM because the status of a node changed from NORMAL to NORMAL.......
However, when we lost 90% of our data center to over heating from a total A/C failure......well, they just assumed an outage that big was a scheduled one they were not made aware of and ignored it, due to it filling their screen up......
All the NMS software in the world can't fix lazy, stupid or overwhelmed.
I have not used whatsup for years and even then, just a demo. I liked hyperic for the free version and in some environments, especially the OpenView and Tivoli ones, the $800 per 2 CPU would be about an 70% discount.
There's a free version of the software. You lose some things like forming groups, etc but it is much better than "whatsup" as it can go on a per service/script scan. The Free version also loses out on the "remediation," as such that if you have a condition like "If apache is not up, run this cleanup and then start apache, and notify this group" Also, Hyperic can give you discounts and even charges half for non production environments.
The non enterprise can be used in a 30 server environment, but you DO give up a lot of functionality and have to re-invent things for it to work. I just wish they open sourced the lot of it, but unlikely
Get Hyperic or one of the other "Small 4" monitoring apps. Never look back.
Long live Hyperic. Free and extensible without the nonsense to set up that is Nagios. You can use Nagios plugins for it if you so wish.
Do not Taunt? Then how will we be able to switch tanks in Phase 2?
He is re-approaching laughing-stock on the way back up to failure.
I am intrigued. Where is the best "XXX" I can buy?
You could always try to see if you could be an IT recruiter. That way, you could be on of the people that vets candidates for jobs and get a tidy little commission out of it. The benefit of it is that you can see what jobs are on the market in your field, as well as making sure that incompetent people have a harder time entering it.
Light bondage requires LC/LC fiber optic cables. You are thinking of copper bondage. ;)
Is it "Talk like an Ass-Pirate Day" already? Wow thailor, I mutht've forgotten all about it, you thilly gooth.