Cheetah2 connection manager and failover arbitrator

IDS 11.50.UC1 Red Hat Linux 5 Simple HDR environment 1 Primary - 1 Secondary I'm taking a look at the connection manager and failover arbitrator shipped with the CSDK in IDS 11.50 and am wondering if there is a better way to recover from a Primary server failure that doesn't require a full reinitialization of HDR. If cheetah01 is primary and cheetah02 is secondary and I onmode -ky cheetah01, the failover arbitrator correctly changes cheetah02's state to Online (Primary) and redirects requests for my primary server SLA to cheetah02. If I want to bring cheetah01 back online as a secondary, is the right thing to do: oninit -PHY onmode -d secondary cheetah02 This appears to work just fine but is there a better, possibly more supported way of doing this? Thanks, Andrew

Andrew Ford wrote: > IDS 11.50.UC1 > Red Hat Linux 5 > Simple HDR environment 1 Primary - 1 Secondary > > I'm taking a look at the connection manager and failover arbitrator shipped with the CSDK in IDS 11.50 and am wondering if there is > a better way to recover from a Primary server failure that doesn't require a full reinitialization of HDR. > > If cheetah01 is primary and cheetah02 is secondary and I onmode -ky cheetah01, the failover arbitrator correctly changes cheetah02's > state to Online (Primary) and redirects requests for my primary server SLA to cheetah02. > > If I want to bring cheetah01 back online as a secondary, is the right thing to do: > > oninit -PHY > onmode -d secondary cheetah02> > This appears to work just fine but is there a better, possibly more supported way of doing this? > > Thanks, > > Andrew Please excuse my ignorance as I have yet to look into it ... but ... is this what is intended to compete with Oracle RAC? -- Daniel A. Morgan University of Washington damorgan@x.washington.edu (replace x with u to respond)

Non-sequitor! No. HDR is far superior to anything that Oracle has, but that may not be to what you refer. Perhaps it was the reference to RSS or Remote Secondary Servers? No to that also. These are additional remote backup servers with their own disk, like a standard HDR secondary but they use a different communications protocol which is more forgiving of longer slower network connections than HDR needs to be. These also have no competition from anything Oracle fields. What you are likely thinking about is SDSS or Shared Disk Secondary Servers which share disk with a standalone server, HDR primary or secondary, and RSS server or even other, multiple, SDSS servers. This feature is the one that competes directly with Oracle RAC. And it fairs pretty well, too. I would, however, like to acknowledge that in this post Daniel has finally admitted his ignorace. This is note worthy and praise worthy! > > > Please excuse my ignorance as I have yet to look into it ... but ... is > this what is intended to compete with Oracle RAC? > -- > Daniel A. Morgan > University of Washington > damorgan@x.washington.edu (replace x with u to respond) > _______________________________________________ > Informix-list mailing list > Informix-list@iiug.org > http://www.iiug.org/mailman/listinfo/informix-list > -- Art S. Kagel Oninit (www.oninit.com) IIUG Board of Directors (art@iiug.org) Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Oninit, the IIUG, nor any other organization with which I am associated either explicitly or implicitly. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves.

Andrew Ford wrote: > IDS 11.50.UC1 > Red Hat Linux 5 > Simple HDR environment 1 Primary - 1 Secondary > > I'm taking a look at the connection manager and failover arbitrator shipped with the CSDK in IDS 11.50 and am wondering if there is > a better way to recover from a Primary server failure that doesn't require a full reinitialization of HDR. > > If cheetah01 is primary and cheetah02 is secondary and I onmode -ky cheetah01, the failover arbitrator correctly changes cheetah02's > state to Online (Primary) and redirects requests for my primary server SLA to cheetah02. > > If I want to bring cheetah01 back online as a secondary, is the right thing to do: > > oninit -PHY > onmode -d secondary cheetah02> > This appears to work just fine but is there a better, possibly more supported way of doing this? > > Thanks, > > Andrew > > There are two scripts (hdrmksec.sh and hdrmkpri.sh) that do this... So, to my best knowledge, this should be supported... Any way, maybe Madison will read this :) Regards,

DA Morgan wrote: > Andrew Ford wrote: >> IDS 11.50.UC1 >> Red Hat Linux 5 >> Simple HDR environment 1 Primary - 1 Secondary >> >> I'm taking a look at the connection manager and failover arbitrator >> shipped with the CSDK in IDS 11.50 and am wondering if there is a >> better way to recover from a Primary server failure that doesn't >> require a full reinitialization of HDR. >> >> If cheetah01 is primary and cheetah02 is secondary and I onmode -ky >> cheetah01, the failover arbitrator correctly changes cheetah02's state >> to Online (Primary) and redirects requests for my primary server SLA >> to cheetah02. >> >> If I want to bring cheetah01 back online as a secondary, is the right >> thing to do: >> >> oninit -PHY >> onmode -d secondary cheetah02>> >> This appears to work just fine but is there a better, possibly more >> supported way of doing this? >> >> Thanks, >> >> Andrew > > Please excuse my ignorance as I have yet to look into it ... but ... is > this what is intended to compete with Oracle RAC? No... The technologies are very different... No, you maybe thinking about writes on secondaries... No, you maybe thinking about enterprise replication... No, this can be used for DR... No, this doesn't cost the same as RAC... No, this works for nodes in different locations... No, this is not new (except for the connection manager, but for this purpose you could use < 11.50... well... No. Regards.

> Andrew Ford wrote: >> IDS 11.50.UC1 >> Red Hat Linux 5 >> Simple HDR environment 1 Primary - 1 Secondary >> >> I'm taking a look at the connection manager and failover arbitrator shipped with the CSDK in IDS 11.50 and am wondering if there >> is >> a better way to recover from a Primary server failure that doesn't require a full reinitialization of HDR. >> >> If cheetah01 is primary and cheetah02 is secondary and I onmode -ky cheetah01, the failover arbitrator correctly changes >> cheetah02's >> state to Online (Primary) and redirects requests for my primary server SLA to cheetah02. >> >> If I want to bring cheetah01 back online as a secondary, is the right thing to do: >> >> oninit -PHY >> onmode -d secondary cheetah02>> >> This appears to work just fine but is there a better, possibly more supported way of doing this? >> >> Thanks, >> >> Andrew >> >> > > There are two scripts (hdrmksec.sh and hdrmkpri.sh) that do this... So, to my > best knowledge, this should be supported... Any way, maybe Madison will read > this :) > > Regards, > _______________________________________________ This excerpt from the hdrmksec.sh script is what had me concerned echo " Note : -PHY is an undocumented option only used in this" echo " HDR failover script" I have no problems using oninit -PHY, we use the hdrmk*.sh scripts all of the time without problems. I'm wondering if there is a new option in 11.50, maybe something a little more documented. :) Thanks, Andrew

Andrew Ford wrote: >> Andrew Ford wrote: >>> IDS 11.50.UC1 >>> Red Hat Linux 5 >>> Simple HDR environment 1 Primary - 1 Secondary >>> >>> I'm taking a look at the connection manager and failover arbitrator shipped with the CSDK in IDS 11.50 and am wondering if there >>> is >>> a better way to recover from a Primary server failure that doesn't require a full reinitialization of HDR. >>> >>> If cheetah01 is primary and cheetah02 is secondary and I onmode -ky cheetah01, the failover arbitrator correctly changes >>> cheetah02's >>> state to Online (Primary) and redirects requests for my primary server SLA to cheetah02. >>> >>> If I want to bring cheetah01 back online as a secondary, is the right thing to do: >>> >>> oninit -PHY >>> onmode -d secondary cheetah02>>> >>> This appears to work just fine but is there a better, possibly more supported way of doing this? >>> >>> Thanks, >>> >>> Andrew >>> >>> >> There are two scripts (hdrmksec.sh and hdrmkpri.sh) that do this... So, to my >> best knowledge, this should be supported... Any way, maybe Madison will read >> this :) >> >> Regards, >> _______________________________________________ > > > > This excerpt from the hdrmksec.sh script is what had me concerned > > echo " Note : -PHY is an undocumented option only used in this" > echo " HDR failover script" > > I have no problems using oninit -PHY, we use the hdrmk*.sh scripts all of the time without problems. I'm wondering if there is a > new option in 11.50, maybe something a little more documented. :) > > Thanks, > > Andrew > > I understand your concern... My point was that if it is included in a script it should be OK. But maybe someone can follow up. Oh... I've found a reference to it in the docs, but it's not a proper documentation of the option: http://publib.boulder.ibm.com/infocenter/idshelp/v115/topic/com.ibm.admin.doc/admin14.htm?resultof=%22%6f%6e%69%6e%69%74%20%2d%70%68%79%22%20 Regards,

Andrew Ford wrote: > > > > This excerpt from the hdrmksec.sh script is what had me concerned > > echo " Note : -PHY is an undocumented option only used in this" > echo " HDR failover script" > > I have no problems using oninit -PHY, we use the hdrmk*.sh scripts all of the time without problems. I'm wondering if there is a > new option in 11.50, maybe something a little more documented. :) > > Thanks, > > Andrew > > Valid point. So I guess I'll document oninit -PHY now.... ;-) oninit -PHY brings the server up as of the last checkpoint. We referto this a physical recovery. Normally when the engine is brought up, we perform physical recovery and that is followed by logical recovery. Logical recovery means that we apply the changes from the logical log file. When we perform oninit -PHY, we are telling the engine to not perform the second phase of recovery (logical recovery) because we are going to perform logical recovery in conjunction with something else. It might be that we want to coordinate the recovery with some primary because the server is an HDR or RSS secondary. In that case we would proceed the logical recovery phase by running onmode -d secondary <HDR_PRIMARY>, or by running onmode -d RSS <primary>. oninit -PHY is very similar in effect to performing a physical restore from a backup (i.e. ontape -p)

Art Kagel wrote: > Non-sequitor! > > No. HDR is far superior to anything that Oracle has, but that may not > be to what you refer. Perhaps it was the reference to RSS or Remote > Secondary Servers? No to that also. These are additional remote backup > servers with their own disk, like a standard HDR secondary but they use > a different communications protocol which is more forgiving of longer > slower network connections than HDR needs to be. These also have no > competition from anything Oracle fields. > > What you are likely thinking about is SDSS or Shared Disk Secondary > Servers which share disk with a standalone server, HDR primary or > secondary, and RSS server or even other, multiple, SDSS servers. This > feature is the one that competes directly with Oracle RAC. And it fairs > pretty well, too. > > I would, however, like to acknowledge that in this post Daniel has > finally admitted his ignorace. This is note worthy and praise worthy! > > Please excuse my ignorance as I have yet to look into it ... but ... is > this what is intended to compete with Oracle RAC? > -- > Daniel A. Morgan > University of Washington > damorgan@x.washington.edu <mailto:damorgan@x.washington.edu> > (replace x with u to respond) > _______________________________________________ > Informix-list mailing list > Informix-list@iiug.org <mailto:Informix-list@iiug.org> > http://www.iiug.org/mailman/listinfo/informix-list > > -- > Art S. Kagel > Oninit (www.oninit.com <http://www.oninit.com>) > IIUG Board of Directors (art@iiug.org <mailto:art@iiug.org>) I wasn't soliciting a marketing diatribe. I was asking a simple Boolean question. Apparently you couldn't answer it. What I am asking about is: 1. Incremental horizontal scaling 2. Transparent failover You addressed neither. -- Daniel A. Morgan University of Washington damorgan@x.washington.edu (replace x with u to respond)

Fernando Nunes wrote: > DA Morgan wrote: >> Andrew Ford wrote: >>> IDS 11.50.UC1 >>> Red Hat Linux 5 >>> Simple HDR environment 1 Primary - 1 Secondary >>> >>> I'm taking a look at the connection manager and failover arbitrator >>> shipped with the CSDK in IDS 11.50 and am wondering if there is a >>> better way to recover from a Primary server failure that doesn't >>> require a full reinitialization of HDR. >>> >>> If cheetah01 is primary and cheetah02 is secondary and I onmode -ky >>> cheetah01, the failover arbitrator correctly changes cheetah02's >>> state to Online (Primary) and redirects requests for my primary >>> server SLA to cheetah02. >>> >>> If I want to bring cheetah01 back online as a secondary, is the right >>> thing to do: >>> >>> oninit -PHY >>> onmode -d secondary cheetah02>>> >>> This appears to work just fine but is there a better, possibly more >>> supported way of doing this? >>> >>> Thanks, >>> >>> Andrew >> >> Please excuse my ignorance as I have yet to look into it ... but ... is >> this what is intended to compete with Oracle RAC? > > No... The technologies are very different... No, you maybe thinking > about writes on secondaries... No, you maybe thinking about enterprise > replication... No, this can be used for DR... No, this doesn't cost the > same as RAC... No, this works for nodes in different locations... No, > this is not new (except for the connection manager, but for this purpose > you could use < 11.50... well... No. > > Regards. Someone recently said that Cheetah 2 had an equivalent to RAC. I am asking whether this is what they were referring to. If not which feature was that statement referring to as I want to set it up in my lab. Thanks. -- Daniel A. Morgan University of Washington damorgan@x.washington.edu (replace x with u to respond)

DA Morgan wrote: > Fernando Nunes wrote: >> DA Morgan wrote: >>> Andrew Ford wrote: >>>> IDS 11.50.UC1 >>>> Red Hat Linux 5 >>>> Simple HDR environment 1 Primary - 1 Secondary >>>> >>>> I'm taking a look at the connection manager and failover arbitrator >>>> shipped with the CSDK in IDS 11.50 and am wondering if there is a >>>> better way to recover from a Primary server failure that doesn't >>>> require a full reinitialization of HDR. >>>> >>>> If cheetah01 is primary and cheetah02 is secondary and I onmode -ky >>>> cheetah01, the failover arbitrator correctly changes cheetah02's >>>> state to Online (Primary) and redirects requests for my primary >>>> server SLA to cheetah02. >>>> >>>> If I want to bring cheetah01 back online as a secondary, is the >>>> right thing to do: >>>> >>>> oninit -PHY >>>> onmode -d secondary cheetah02>>>> >>>> This appears to work just fine but is there a better, possibly more >>>> supported way of doing this? >>>> >>>> Thanks, >>>> >>>> Andrew >>> >>> Please excuse my ignorance as I have yet to look into it ... but ... is >>> this what is intended to compete with Oracle RAC? >> >> No... The technologies are very different... No, you maybe thinking >> about writes on secondaries... No, you maybe thinking about enterprise >> replication... No, this can be used for DR... No, this doesn't cost >> the same as RAC... No, this works for nodes in different locations... >> No, this is not new (except for the connection manager, but for this >> purpose you could use < 11.50... well... No. >> >> Regards. > > Someone recently said that Cheetah 2 had an equivalent to RAC. I am > asking whether this is what they were referring to. If not which > feature was that statement referring to as I want to set it up in > my lab. Thanks. You should ask the person who said so. And it's nice to see that someone with so little knowledge about IDS can set this up in the lab. It really means it's simple to setup. Last time I talked to someone attending a RAC training they spent almost half of the training installing patches just to install/configure it... :) Regards.

Fernando Nunes wrote: > DA Morgan wrote: >> Fernando Nunes wrote: >>> DA Morgan wrote: >>>> Andrew Ford wrote: >>>>> IDS 11.50.UC1 >>>>> Red Hat Linux 5 >>>>> Simple HDR environment 1 Primary - 1 Secondary >>>>> >>>>> I'm taking a look at the connection manager and failover arbitrator >>>>> shipped with the CSDK in IDS 11.50 and am wondering if there is a >>>>> better way to recover from a Primary server failure that doesn't >>>>> require a full reinitialization of HDR. >>>>> >>>>> If cheetah01 is primary and cheetah02 is secondary and I onmode -ky >>>>> cheetah01, the failover arbitrator correctly changes cheetah02's >>>>> state to Online (Primary) and redirects requests for my primary >>>>> server SLA to cheetah02. >>>>> >>>>> If I want to bring cheetah01 back online as a secondary, is the >>>>> right thing to do: >>>>> >>>>> oninit -PHY >>>>> onmode -d secondary cheetah02>>>>> >>>>> This appears to work just fine but is there a better, possibly more >>>>> supported way of doing this? >>>>> >>>>> Thanks, >>>>> >>>>> Andrew >>>> >>>> Please excuse my ignorance as I have yet to look into it ... but ... is >>>> this what is intended to compete with Oracle RAC? >>> >>> No... The technologies are very different... No, you maybe thinking >>> about writes on secondaries... No, you maybe thinking about >>> enterprise replication... No, this can be used for DR... No, this >>> doesn't cost the same as RAC... No, this works for nodes in different >>> locations... No, this is not new (except for the connection manager, >>> but for this purpose you could use < 11.50... well... No. >>> >>> Regards. >> >> Someone recently said that Cheetah 2 had an equivalent to RAC. I am >> asking whether this is what they were referring to. If not which >> feature was that statement referring to as I want to set it up in >> my lab. Thanks. > > You should ask the person who said so. And it's nice to see that someone > with so little knowledge about IDS can set this up in the lab. It really > means it's simple to setup. Last time I talked to someone attending a > RAC training they spent almost half of the training installing patches > just to install/configure it... :) > Regards. Yeah. Unlike Oracle which is really difficult to install as demonstrated here: http://www.youtube.com/watch?v=CHzV4LZnvHc and here: http://www.youtube.com/watch?v=SNW-7jCBWsw One of these days some of you guys are going to wake up and discover it is no longer the 80s. So the conclusion is that if Cheetah 2 has a feature that does what RAC does not a single person here can name it. Amazing! -- Daniel A. Morgan University of Washington damorgan@x.washington.edu (replace x with u to respond)

See below. On Sun, Jun 8, 2008 at 1:29 PM, DA Morgan <damorgan@psoug.org> wrote: > <SNIP> > >> Someone recently said that Cheetah 2 had an equivalent to RAC. I am > >> asking whether this is what they were referring to. If not which > >> feature was that statement referring to as I want to set it up in > >> my lab. Thanks. > > > > You should ask the person who said so. And it's nice to see that someone > > with so little knowledge about IDS can set this up in the lab. It really > > means it's simple to setup. Last time I talked to someone attending a > > RAC training they spent almost half of the training installing patches > > just to install/configure it... :) > > Regards. > > Yeah. Unlike Oracle which is really difficult to install as > demonstrated here: > http://www.youtube.com/watch?v=CHzV4LZnvHc > and here: > http://www.youtube.com/watch?v=SNW-7jCBWsw > > One of these days some of you guys are going to wake up > and discover it is no longer the 80s. > > So the conclusion is that if Cheetah 2 has a feature that does > what RAC does not a single person here can name it. Amazing! > You cannot be this dense and still hold a chair at a major University! Among other posts that I saw, I said Informix Dynamic Server Shared Disk Secondary Server (SDSS) feature in IDS 11.50 is functionally equivalent to Oracle RAC, does everything that RAC does and does it better. It is so trivial to set up that you just have to start the SDSS instance with the correct commandline argument so that the newly started server instance does not try to grab exclusive access to the disks at startup, and then run a simple commandline utility to tell both the primary server and the new secondary to begin handshaking. SDSS servers can reside on the same physical machine as the primary to provide private cache for reports so they do not affect production caching, on colocated servers and blades that share the same physical disk farm, or on remote servers using real-time disk array level replication of the drives over sufficiently high speed connections. POOF! As for that silly video that perports to show how simple and quick it is to install and start an Oracle instance, we in the Informix community laughed for days over that one! You (and I do mean you Daniel) can install, configure, and start at least six instances sequentially in the four plus hours that were compressed into that video. Probably an experienced DBA using the commandline installer on any UNIX box could install more than four in the real-time two hours the video actually plays for. It took me 40 mins to install IDS on my laptop including 3 demo databases, separate data and log dbspaces, and automated logical log archiving, and that's on Vista where I cannot use the commandline installer and it's the first Windows installation I've ever performed - let's see 40 * 6 = 240 minutes, yup 4 hours. OMG ROTFL. > > _______________________________________________ > Informix-list mailing list > Informix-list@iiug.org > http://www.iiug.org/mailman/listinfo/informix-list > -- Art S. Kagel Oninit (www.oninit.com) IIUG Board of Directors (art@iiug.org) Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Oninit, the IIUG, nor any other organization with which I am associated either explicitly or implicitly. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves.

Daniel, I think u are getting valid responses based on who you are and how you treat people in this forum. Isn't your posting here as an oracle expert expecting us to know the atomic features of an oracle solution a bit like an informix expert posting in an oracle forum asking if there are features of oracle that atomically compete with features of cheetah, without defining in detail what feature they imply. While there are some pros who know exactly what u might mean by "oracle rac", perhaps the majority only know the buzzword, not the underlying meaning/architecture of it. You have also not exactly endeared yourself to this community, which will impact the friendliness of answers you may receive. Would you insult Tom Kyte and then expect a valid answer? Norma Jean . . . -----Original Message----- From: informix-list-bounces@iiug.org [mailto:informix-list-bounces@iiug.org] On Behalf Of DA Morgan Sent: Sunday, June 08, 2008 12:29 PM To: informix-list@iiug.org Subject: Re: Cheetah2 connection manager and failover arbitrator [snip].... So the conclusion is that if Cheetah 2 has a feature that does what RAC does not a single person here can name it. Amazing! -- Daniel A. Morgan University of Washington damorgan@x.washington.edu (replace x with u to respond) _______________________________________________ Informix-list mailing list Informix-list@iiug.org http://www.iiug.org/mailman/listinfo/informix-list ============================================================ The information contained in this message may be privileged and confidential and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any reproduction, dissemination or distribution of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. Thank you. Tellabs ============================================================

DA Morgan wrote: > Fernando Nunes wrote: >> DA Morgan wrote: >>> Fernando Nunes wrote: >>>> DA Morgan wrote: >>>>> Andrew Ford wrote: >>>>>> IDS 11.50.UC1 >>>>>> Red Hat Linux 5 >>>>>> Simple HDR environment 1 Primary - 1 Secondary >>>>>> >>>>>> I'm taking a look at the connection manager and failover >>>>>> arbitrator shipped with the CSDK in IDS 11.50 and am wondering if >>>>>> there is a better way to recover from a Primary server failure >>>>>> that doesn't require a full reinitialization of HDR. >>>>>> >>>>>> If cheetah01 is primary and cheetah02 is secondary and I onmode >>>>>> -ky cheetah01, the failover arbitrator correctly changes >>>>>> cheetah02's state to Online (Primary) and redirects requests for >>>>>> my primary server SLA to cheetah02. >>>>>> >>>>>> If I want to bring cheetah01 back online as a secondary, is the >>>>>> right thing to do: >>>>>> >>>>>> oninit -PHY >>>>>> onmode -d secondary cheetah02>>>>>> >>>>>> This appears to work just fine but is there a better, possibly >>>>>> more supported way of doing this? >>>>>> >>>>>> Thanks, >>>>>> >>>>>> Andrew >>>>> >>>>> Please excuse my ignorance as I have yet to look into it ... but >>>>> ... is >>>>> this what is intended to compete with Oracle RAC? >>>> >>>> No... The technologies are very different... No, you maybe thinking >>>> about writes on secondaries... No, you maybe thinking about >>>> enterprise replication... No, this can be used for DR... No, this >>>> doesn't cost the same as RAC... No, this works for nodes in >>>> different locations... No, this is not new (except for the >>>> connection manager, but for this purpose you could use < 11.50... >>>> well... No. >>>> >>>> Regards. >>> >>> Someone recently said that Cheetah 2 had an equivalent to RAC. I am >>> asking whether this is what they were referring to. If not which >>> feature was that statement referring to as I want to set it up in >>> my lab. Thanks. >> >> You should ask the person who said so. And it's nice to see that >> someone with so little knowledge about IDS can set this up in the lab. >> It really means it's simple to setup. Last time I talked to someone >> attending a RAC training they spent almost half of the training >> installing patches just to install/configure it... :) >> Regards. > > Yeah. Unlike Oracle which is really difficult to install as > demonstrated here: > http://www.youtube.com/watch?v=CHzV4LZnvHc > and here: > http://www.youtube.com/watch?v=SNW-7jCBWsw > > One of these days some of you guys are going to wake up > and discover it is no longer the 80s. > > So the conclusion is that if Cheetah 2 has a feature that does > what RAC does not a single person here can name it. Amazing! If you want to discover the new features of Cheetah 2 you can look at the docs, download it, listen to the webcasts or register yourself to one of the several practical workshops going around the world these days. I can't think about a single reason why I should make your life easier... Besides this, you could at least read the answers that were given to you. I don't know about any feature that "does what RAC does" simply because I don't really care about it, or about you. I can talk about a lot of features that work for solving customer problems... I'm not sure if RAC can be used for this. We just have to read the Oracle newsgroups to understand why Oracle sells it (and not why customers buy it). Regards.

Art Kagel wrote: > See below. > > > On Sun, Jun 8, 2008 at 1:29 PM, DA Morgan <damorgan@psoug.org > <mailto:damorgan@psoug.org>> wrote: > > <SNIP> > >> Someone recently said that Cheetah 2 had an equivalent to RAC. I am > >> asking whether this is what they were referring to. If not which > >> feature was that statement referring to as I want to set it up in > >> my lab. Thanks. > > > > You should ask the person who said so. And it's nice to see that > someone > > with so little knowledge about IDS can set this up in the lab. It > really > > means it's simple to setup. Last time I talked to someone attending a > > RAC training they spent almost half of the training installing > patches > > just to install/configure it... :) > > Regards. > > Yeah. Unlike Oracle which is really difficult to install as > demonstrated here: > http://www.youtube.com/watch?v=CHzV4LZnvHc > and here: > http://www.youtube.com/watch?v=SNW-7jCBWsw > > One of these days some of you guys are going to wake up > and discover it is no longer the 80s. > > So the conclusion is that if Cheetah 2 has a feature that does > what RAC does not a single person here can name it. Amazing! > > > You cannot be this dense and still hold a chair at a major University! > > Among other posts that I saw, I said Informix Dynamic Server Shared Disk > Secondary Server (SDSS) feature in IDS 11.50 is functionally equivalent > to Oracle RAC, does everything that RAC does and does it better. It is > so trivial to set up that you just have to start the SDSS instance with > the correct commandline argument so that the newly started server > instance does not try to grab exclusive access to the disks at startup, > and then run a simple commandline utility to tell both the primary > server and the new secondary to begin handshaking. SDSS servers can > reside on the same physical machine as the primary to provide private > cache for reports so they do not affect production caching, on colocated > servers and blades that share the same physical disk farm, or on remote > servers using real-time disk array level replication of the drives over > sufficiently high speed connections. POOF! > > As for that silly video that perports to show how simple and quick it is > to install and start an Oracle instance, we in the Informix community > laughed for days over that one! You (and I do mean you Daniel) can > install, configure, and start at least six instances sequentially in the > four plus hours that were compressed into that video. Probably an > experienced DBA using the commandline installer on any UNIX box could > install more than four in the real-time two hours the video actually > plays for. It took me 40 mins to install IDS on my laptop including 3 > demo databases, separate data and log dbspaces, and automated logical > log archiving, and that's on Vista where I cannot use the commandline > installer and it's the first Windows installation I've ever performed - > let's see 40 * 6 = 240 minutes, yup 4 hours. OMG ROTFL. > > > > _______________________________________________ > Informix-list mailing list > Informix-list@iiug.org <mailto:Informix-list@iiug.org> > http://www.iiug.org/mailman/listinfo/informix-list > > -- > Art S. Kagel > Oninit (www.oninit.com <http://www.oninit.com>) > IIUG Board of Directors (art@iiug.org <mailto:art@iiug.org>) > > Disclaimer: Please keep in mind that my own opinions are my own opinions > and do not reflect on my employer, Oninit, the IIUG, nor any other > organization with which I am associated either explicitly or implicitly. > Neither do those opinions reflect those of other individuals affiliated > with any entity with which I am affiliated nor those of the entities > themselves. Dense? I've read the Informix docs on DSSD. http://www.redbooks.ibm.com/redpieces/abstracts/SG247488.html?Open If you think this does what RAC does ... what you are primarily advertising is your ignorance about RAC so you should be careful about throwing stones. You may not like the fact that I think Informix has all the market influence of Rdb and Pick. But give me some credit for reading the docs before asking this bunch for help. I thought, perhaps, RAC had some actual competition which would have been a good thing ... this just isn't it. -- Daniel A. Morgan University of Washington damorgan@x.washington.edu (replace x with u to respond)

Fernando Nunes wrote: > DA Morgan wrote: >> Fernando Nunes wrote: >>> DA Morgan wrote: >>>> Fernando Nunes wrote: >>>>> DA Morgan wrote: >>>>>> Andrew Ford wrote: >>>>>>> IDS 11.50.UC1 >>>>>>> Red Hat Linux 5 >>>>>>> Simple HDR environment 1 Primary - 1 Secondary >>>>>>> >>>>>>> I'm taking a look at the connection manager and failover >>>>>>> arbitrator shipped with the CSDK in IDS 11.50 and am wondering if >>>>>>> there is a better way to recover from a Primary server failure >>>>>>> that doesn't require a full reinitialization of HDR. >>>>>>> >>>>>>> If cheetah01 is primary and cheetah02 is secondary and I onmode >>>>>>> -ky cheetah01, the failover arbitrator correctly changes >>>>>>> cheetah02's state to Online (Primary) and redirects requests for >>>>>>> my primary server SLA to cheetah02. >>>>>>> >>>>>>> If I want to bring cheetah01 back online as a secondary, is the >>>>>>> right thing to do: >>>>>>> >>>>>>> oninit -PHY >>>>>>> onmode -d secondary cheetah02>>>>>>> >>>>>>> This appears to work just fine but is there a better, possibly >>>>>>> more supported way of doing this? >>>>>>> >>>>>>> Thanks, >>>>>>> >>>>>>> Andrew >>>>>> >>>>>> Please excuse my ignorance as I have yet to look into it ... but >>>>>> ... is >>>>>> this what is intended to compete with Oracle RAC? >>>>> >>>>> No... The technologies are very different... No, you maybe thinking >>>>> about writes on secondaries... No, you maybe thinking about >>>>> enterprise replication... No, this can be used for DR... No, this >>>>> doesn't cost the same as RAC... No, this works for nodes in >>>>> different locations... No, this is not new (except for the >>>>> connection manager, but for this purpose you could use < 11.50... >>>>> well... No. >>>>> >>>>> Regards. >>>> >>>> Someone recently said that Cheetah 2 had an equivalent to RAC. I am >>>> asking whether this is what they were referring to. If not which >>>> feature was that statement referring to as I want to set it up in >>>> my lab. Thanks. >>> >>> You should ask the person who said so. And it's nice to see that >>> someone with so little knowledge about IDS can set this up in the >>> lab. It really means it's simple to setup. Last time I talked to >>> someone attending a RAC training they spent almost half of the >>> training installing patches just to install/configure it... :) >>> Regards. >> >> Yeah. Unlike Oracle which is really difficult to install as >> demonstrated here: >> http://www.youtube.com/watch?v=CHzV4LZnvHc >> and here: >> http://www.youtube.com/watch?v=SNW-7jCBWsw >> >> One of these days some of you guys are going to wake up >> and discover it is no longer the 80s. >> >> So the conclusion is that if Cheetah 2 has a feature that does >> what RAC does not a single person here can name it. Amazing! > > If you want to discover the new features of Cheetah 2 you can look at > the docs, download it, listen to the webcasts or register yourself to > one of the several practical workshops going around the world these > days. I can't think about a single reason why I should make your life > easier... > > Besides this, you could at least read the answers that were given to > you. I don't know about any feature that "does what RAC does" simply > because I don't really care about it, or about you. I can talk about a > lot of features that work for solving customer problems... I'm not sure > if RAC can be used for this. We just have to read the Oracle newsgroups > to understand why Oracle sells it (and not why customers buy it). > > Regards. I have. I regularly read the Redbook docs such as SG24-7488-01. But I couldn't find anything that did what RAC does. Given Art's comments it appears that the new featuers, while nice, still do not duplicate the Oracle capability. I would suggest the Oracle newsgroup is the last place to go for getting an in depth understanding of RAC. If you truly want to understand it the places to go are: http://www.oracleracsig.org/pls/apex/f?p=105:100:2492374266755309 http://www.oracle.com/technology/deploy/availability/htdocs/maa.htm Thanks. -- Daniel A. Morgan Oracle Ace Director & Instructor University of Washington damorgan@x.washington.edu (replace x with u to respond) Puget Sound Oracle Users Group www.psoug.org

DA Morgan said: > I've read the Informix docs on DSSD. You have? I've been working with Informix for 20 years now, and I've never heard of an Informix feature called DSSD. > http://www.redbooks.ibm.com/redpieces/abstracts/SG247488.html?Open > > If you think this does what RAC does ... what you are primarily > advertising is your ignorance about RAC so you should be careful > about throwing stones. Actually, what I think RAC does is destroy scalability and eat up administration resources for a half-baked, badly-thought-out bit of technology that only an idiot could admir... oh. > I thought, perhaps, RAC had some actual competition which would have > been a good thing ... this just isn't it. No, Informix doesn't do stuff that fucks scalability ... they leave that to the professional fuckers of scalability, the guys you so impartially root for. ;o) -- Bye now, Obnoxio "There were a myriad of problems which conspired to corrupt your reason and rob you of your common sense. Fear got the best of you, and in your panic you turned to the Labour Party. They promised you order, they promised you peace, and all they demanded in return was your silent, obedient consent."

Sebastian, Norma J. wrote: > Daniel, > I think u are getting valid responses based on who you are and how you > treat people in this forum. Isn't your posting here as an oracle expert > expecting us to know the atomic features of an oracle solution a bit > like an informix expert posting in an oracle forum asking if there are > features of oracle that atomically compete with features of cheetah, > without defining in detail what feature they imply. While there are > some pros who know exactly what u might mean by "oracle rac", perhaps > the majority only know the buzzword, not the underlying > meaning/architecture of it. You have also not exactly endeared yourself > to this community, which will impact the friendliness of answers you may > receive. Would you insult Tom Kyte and then expect a valid answer? > Norma Jean I don't think your read on this is correct. The statement was made in this forum that Cheetah 2 had the capabilities of RAC. That statement presumed that the person that wrote it knew enough about RAC to make the comparison. Thus downloaded a lot of Informix docs looking for the new feature that might, one would hope, push Oracle to improve more to keep its competitive edge. Alas I couldn't find it so I asked for help. Personally I think Informix was (past tense) the best RDBMS on the market. I think it was trashed by the C-Level management of Informix the company and is dying of benign neglect on the part of IBM. That is not a negative statement about the product or those that use it. That is a negative statement about the quality of the management that turned a once great product into a basket case and led to statistics such as these: www.dice.com (08-Jun-2008) Oracle 17,978 listed jobs SQL Server 13,961 DB2 2,744 Sybase 1,583 Pick 318 Informix 284 You may not like the messenger but the statistics are what matters. -- Daniel A. Morgan University of Washington damorgan@x.washington.edu (replace x with u to respond)

DA Morgan wrote: > Fernando Nunes wrote: >> DA Morgan wrote: >>> Fernando Nunes wrote: >>>> DA Morgan wrote: >>>>> Fernando Nunes wrote: >>>>>> DA Morgan wrote: >>>>>>> Andrew Ford wrote: >>>>>>>> IDS 11.50.UC1 >>>>>>>> Red Hat Linux 5 >>>>>>>> Simple HDR environment 1 Primary - 1 Secondary >>>>>>>> >>>>>>>> I'm taking a look at the connection manager and failover >>>>>>>> arbitrator shipped with the CSDK in IDS 11.50 and am wondering >>>>>>>> if there is a better way to recover from a Primary server >>>>>>>> failure that doesn't require a full reinitialization of HDR. >>>>>>>> >>>>>>>> If cheetah01 is primary and cheetah02 is secondary and I onmode >>>>>>>> -ky cheetah01, the failover arbitrator correctly changes >>>>>>>> cheetah02's state to Online (Primary) and redirects requests for >>>>>>>> my primary server SLA to cheetah02. >>>>>>>> >>>>>>>> If I want to bring cheetah01 back online as a secondary, is the >>>>>>>> right thing to do: >>>>>>>> >>>>>>>> oninit -PHY >>>>>>>> onmode -d secondary cheetah02>>>>>>>> >>>>>>>> This appears to work just fine but is there a better, possibly >>>>>>>> more supported way of doing this? >>>>>>>> >>>>>>>> Thanks, >>>>>>>> >>>>>>>> Andrew >>>>>>> >>>>>>> Please excuse my ignorance as I have yet to look into it ... but >>>>>>> ... is >>>>>>> this what is intended to compete with Oracle RAC? >>>>>> >>>>>> No... The technologies are very different... No, you maybe >>>>>> thinking about writes on secondaries... No, you maybe thinking >>>>>> about enterprise replication... No, this can be used for DR... No, >>>>>> this doesn't cost the same as RAC... No, this works for nodes in >>>>>> different locations... No, this is not new (except for the >>>>>> connection manager, but for this purpose you could use < 11.50... >>>>>> well... No. >>>>>> >>>>>> Regards. >>>>> >>>>> Someone recently said that Cheetah 2 had an equivalent to RAC. I am >>>>> asking whether this is what they were referring to. If not which >>>>> feature was that statement referring to as I want to set it up in >>>>> my lab. Thanks. >>>> >>>> You should ask the person who said so. And it's nice to see that >>>> someone with so little knowledge about IDS can set this up in the >>>> lab. It really means it's simple to setup. Last time I talked to >>>> someone attending a RAC training they spent almost half of the >>>> training installing patches just to install/configure it... :) >>>> Regards. >>> >>> Yeah. Unlike Oracle which is really difficult to install as >>> demonstrated here: >>> http://www.youtube.com/watch?v=CHzV4LZnvHc >>> and here: >>> http://www.youtube.com/watch?v=SNW-7jCBWsw >>> >>> One of these days some of you guys are going to wake up >>> and discover it is no longer the 80s. >>> >>> So the conclusion is that if Cheetah 2 has a feature that does >>> what RAC does not a single person here can name it. Amazing! >> >> If you want to discover the new features of Cheetah 2 you can look at >> the docs, download it, listen to the webcasts or register yourself to >> one of the several practical workshops going around the world these >> days. I can't think about a single reason why I should make your life >> easier... >> >> Besides this, you could at least read the answers that were given to >> you. I don't know about any feature that "does what RAC does" simply >> because I don't really care about it, or about you. I can talk about a >> lot of features that work for solving customer problems... I'm not >> sure if RAC can be used for this. We just have to read the Oracle >> newsgroups to understand why Oracle sells it (and not why customers >> buy it). >> >> Regards. > > I have. I regularly read the Redbook docs such as SG24-7488-01. But I > couldn't find anything that did what RAC does. Given Art's comments > it appears that the new featuers, while nice, still do not duplicate > the Oracle capability. You may want to stop for a second and remember it was you who talked about RAC. I don't talk about RAC. It should be the Oracle solution to some problem(s). As such it doesn't belong here in this group. Here we are supposed to talk about Informix solutions to problems, or about Informix problems. You love to post on c.d.o.s and other groups about off-topic. Would you mind to follow your own advice? Besides, if you're used to read the RedBooks and other Docs why do you start participating in a thread ignoring the OP question and asking if an old feature is the "new feature that does what Oracle does"? Only after getting the answer you deserve (if you deserve any answer) you come back and paste a few links saying "oh... it's not like RAC"? If you really follow up the new features you should know it at the beginning. > I would suggest the Oracle newsgroup is the last place to go for getting > an in depth understanding of RAC. If you truly want to understand it the > places to go are: Yes. I bet it is. On the other hand it's a nice place to understand that customer are buying it for one thing and what they get is a completely different one. Allow me to quote you: " RAC is not a perfect solution for HA but is can be a very good one if your application supports node affinity ... not a lot of sharing of blocks and locks. This is something you should test before you move to implementation. " What the hell? Isn't the "sharing of blocks and locks" the fun part in RAC? If even you suggest that this should be tested before implementation, what is it good for? Note for the rest of the group. This is DA Morgan speaking about Oracle... Please put "not a perfect" and "can be a very good" sentences in perspective :) Regards.

DA Morgan wrote: > Sebastian, Norma J. wrote: >> Daniel, >> I think u are getting valid responses based on who you are and how you >> treat people in this forum. Isn't your posting here as an oracle expert >> expecting us to know the atomic features of an oracle solution a bit >> like an informix expert posting in an oracle forum asking if there are >> features of oracle that atomically compete with features of cheetah, >> without defining in detail what feature they imply. While there are >> some pros who know exactly what u might mean by "oracle rac", perhaps >> the majority only know the buzzword, not the underlying >> meaning/architecture of it. You have also not exactly endeared yourself >> to this community, which will impact the friendliness of answers you may >> receive. Would you insult Tom Kyte and then expect a valid answer? >> Norma Jean > > I don't think your read on this is correct. > > The statement was made in this forum that Cheetah 2 had the capabilities > of RAC. That statement presumed that the person that wrote it knew > enough about RAC to make the comparison. Really? Not in this thread... In this particular thread only you talked about RAC... You should have picked the subject in the correct thread. > > Thus downloaded a lot of Informix docs looking for the new feature > that might, one would hope, push Oracle to improve more to keep its > competitive edge. Alas I couldn't find it so I asked for help. Hmm.... after... as I said in another post. But we all know you're not a usual user of Redbooks :) What we really know is that after talking too much, too quickly and too incorrectly you'll browse the Internet to try to prove your point... But you should take some more of your time to read carefully so that you don't keep saying mistakes. > > Personally I think Informix was (past tense) the best RDBMS on the > market. I think it was trashed by the C-Level management of Informix > the company and is dying of benign neglect on the part of IBM. That > is not a negative statement about the product or those that use it. > That is a negative statement about the quality of the management that > turned a once great product into a basket case and led to statistics > such as these: That was in the 80's you've talked about previously? > www.dice.com (08-Jun-2008) > Oracle 17,978 listed jobs > SQL Server 13,961 > DB2 2,744 > Sybase 1,583 > Pick 318 > Informix 284 > > You may not like the messenger but the statistics are what matters. And there we go again. In a thread about a specific issue with Informix, you just ignore it, talk about RAC, and then go on with the old trick of dice.com. Would you care to show just a little respect for the people who read this newsgroup? Stick to the topics. Your Usenet rules preaching looks good in c.d.o.s and other c.d.o.* groups, but don't serve you here?!

DA Morgan wrote: > Andrew Ford wrote: >> IDS 11.50.UC1 >> Red Hat Linux 5 >> Simple HDR environment 1 Primary - 1 Secondary > > Please excuse my ignorance as I have yet to look into it ... but ... is > this what is intended to compete with Oracle RAC? The connection manager provides a form of connection virtualization when there are multiple nodes on which work can be performed. The most simple usage of the CM would be to manage connectivity with the HDR pair.

On 8 Jun, 23:02, DA Morgan <damor...@psoug.org> wrote: > Fernando Nunes wrote: > > DA Morgan wrote: > >> Fernando Nunes wrote: > >>> DA Morgan wrote: > >>>> Fernando Nunes wrote: > >>>>> DA Morgan wrote: > >>>>>> Andrew Ford wrote: > >>>>>>> IDS 11.50.UC1 > >>>>>>> Red Hat Linux 5 > >>>>>>> Simple HDR environment 1 Primary - 1 Secondary > > >>>>>>> I'm taking a look at the connection manager and failover > >>>>>>> arbitrator shipped with the CSDK in IDS 11.50 and am wondering if > >>>>>>> there is a better way to recover from a Primary server failure > >>>>>>> that doesn't require a full reinitialization of HDR. > > >>>>>>> If cheetah01 is primary and cheetah02 is secondary and I onmode > >>>>>>> -ky cheetah01, the failover arbitrator correctly changes > >>>>>>> cheetah02's state to Online (Primary) and redirects requests for > >>>>>>> my primary server SLA to cheetah02. > > >>>>>>> If I want to bring cheetah01 back online as a secondary, is the > >>>>>>> right thing to do: > > >>>>>>> oninit -PHY > >>>>>>> onmode -d secondary cheetah02> > >>>>>>> This appears to work just fine but is there a better, possibly > >>>>>>> more supported way of doing this? > > >>>>>>> Thanks, > > >>>>>>> Andrew > > >>>>>> Please excuse my ignorance as I have yet to look into it ... but > >>>>>> ... is > >>>>>> this what is intended to compete with Oracle RAC? > > >>>>> No... The technologies are very different... No, you maybe thinking > >>>>> about writes on secondaries... No, you maybe thinking about > >>>>> enterprise replication... No, this can be used for DR... No, this > >>>>> doesn't cost the same as RAC... No, this works for nodes in > >>>>> different locations... No, this is not new (except for the > >>>>> connection manager, but for this purpose you could use < 11.50... > >>>>> well... No. > > >>>>> Regards. > > >>>> Someone recently said that Cheetah 2 had an equivalent to RAC. I am > >>>> asking whether this is what they were referring to. If not which > >>>> feature was that statement referring to as I want to set it up in > >>>> my lab. Thanks. > > >>> You should ask the person who said so. And it's nice to see that > >>> someone with so little knowledge about IDS can set this up in the > >>> lab. It really means it's simple to setup. Last time I talked to > >>> someone attending a RAC training they spent almost half of the > >>> training installing patches just to install/configure it... :) > >>> Regards. > > >> Yeah. Unlike Oracle which is really difficult to install as > >> demonstrated here: > >>http://www.youtube.com/watch?v=CHzV4LZnvHc > >> and here: > >>http://www.youtube.com/watch?v=SNW-7jCBWsw > > >> One of these days some of you guys are going to wake up > >> and discover it is no longer the 80s. > > >> So the conclusion is that if Cheetah 2 has a feature that does > >> what RAC does not a single person here can name it. Amazing! > > > If you want to discover the new features of Cheetah 2 you can look at > > the docs, download it, listen to the webcasts or register yourself to > > one of the several practical workshops going around the world these > > days. I can't think about a single reason why I should make your life > > easier... > > > Besides this, you could at least read the answers that were given to > > you. I don't know about any feature that "does what RAC does" simply > > because I don't really care about it, or about you. I can talk about a > > lot of features that work for solving customer problems... I'm not sure > > if RAC can be used for this. We just have to read the Oracle newsgroups > > to understand why Oracle sells it (and not why customers buy it). > > > Regards. > > I have. I regularly read the Redbook docs such as SG24-7488-01. But I > couldn't find anything that did what RAC does. Given Art's comments > it appears that the new featuers, while nice, still do not duplicate > the Oracle capability. > > I would suggest the Oracle newsgroup is the last place to go for getting > an in depth understanding of RAC. If you truly want to understand it the > places to go are:http://www.oracleracsig.org/pls/apex/f?p=105:100:2492374266755309http://www.oracle.com/technology/deploy/availability/htdocs/maa.htm > > Thanks. > -- > Daniel A. Morgan > Oracle Ace Director & Instructor > University of Washington > damor...@x.washington.edu (replace x with u to respond) > Puget Sound Oracle Users Groupwww.psoug.org- Hide quoted text - > > - Show quoted text - You can have one server and - several servers sharing the same disks (SDS) - several servers completely replicated from the same server (HDR/RSS) - several servers replicated from each other at the table and column level (ER) or all of these combined. So what does RAC do that Informix does not do?