Re: [SSI] Re: ssic-linux-devel digest, Vol 1 #289 - 5 msgs
Brought to you by:
brucewalker,
rogertsang
Menu
▾
▴
Re: [SSI] Re: ssic-linux-devel digest, Vol 1 #289 - 5 msgs
|
From: Sharad T. <mai...@in...> - 2002-10-28 10:46:51
|
Hi all ... some more questions ... 1) Where is the information of shared hardware and software resources stored ? ( Also if I have to add a new resource at any node for sharing in the whole cluster how do i do that, and where will be this information be stored ? 2) In the absence of CLS, if my primary node goes down, will a secondary node take over or the whole cluster goes down. 3) How does the services differ if I use only CI ? (I say this because I believe I can get SSI benefits only in the presence of CFS or GFS) ... Thanking you all in advance. Cheers!!!! Best Wishes and Warm regads Sharad "Paul Kronenwetter" wrote: Sharad, Be sure that the DHCP server that answers your client is your new SSI PC and points it to the kernel in /tftpboot on your new SSI PC. There are a few other configuration options that are set in the DHCP server too, so be sure to go over that section again. Hope this helps! -Paul ssi...@li... wrote: >Send ssic-linux-devel mailing list submissions to > ssi...@li... > >To subscribe or unsubscribe via the World Wide Web, visit > https://lists.sourceforge.net/lists/listinfo/ssic-linux-devel >or, via email, send a message with subject or body 'help' to > ssi...@li... > >You can reach the person managing the list at > ssi...@li... > >When replying, please edit your Subject line so it is more specific >than "Re: Contents of ssic-linux-devel digest..." > > >Today's Topics: > > 1. Re: Issues with adding new functionality to SSI (Brian J. Watson) > 2. Re: [CI] Changes needed to build the CVS version of CI. (Brian J. Watson) > 3. Re: [CI] Changes needed to build the CVS version of CI. (Aneesh Kumar K.V) > 4. Re: Queries related to SSI Linux (Aneesh Kumar K.V) > 5. problem adding new node (Sharad Tiwari) > >--__--__-- > >Message: 1 >Date: Fri, 25 Oct 2002 14:53:39 -0700 >From: "Brian J. Watson" >To: "Aneesh Kumar K.V" >Cc: ssic-linux-devel , > CI discussion >Subject: Re: [SSI] Issues with adding new functionality to SSI > >"Aneesh Kumar K.V" wrote: > > >>Hi, >> >> Anybody looking into these issues ?. Is there anything wrong in using >>the ics_vproc_chan for ipvs-bind/CVIP related functionality. >> >> >> > >Since Kai's left, there's no ICS expert on the CI or SSI project. I'm >not familiar with the issues of selecting channels. I imagine it's okay >to use ics_vproc_chan for now, then switch to something else in the >future if it's a problem. > >It's probably a bug that you can't easily add another channel. Can you >add it to the SF database? > >-Brian > > >--__--__-- > >Message: 2 >Date: Fri, 25 Oct 2002 14:58:50 -0700 >From: "Brian J. Watson" >To: "Aneesh Kumar K.V" , > Cluster CI , > SSI discussion >Subject: [SSI] Re: [CI] Changes needed to build the CVS version of CI. > >"Aneesh Kumar K.V" wrote: > > >>Hi, >> >> Attaching below is the changes needed to build the CVS version of CI. >>Mostly #ifdef changes. I am not sure i am using the right #ifdef . Can >>someone verify it and do the necessary checkin. ? >> >> > > >I'm a bit torn about the issue of SSI code existing in the CI >repository. Ideally it shouldn't be there. Any CI file touched by SSI >should exist in both repositories, with the SSI changes made to only the >SSI version. > >Although this would be the purest approach, it also creates a bit of a >maintenance headache for CI. Any change made to the CI version of one of >these files must also be made to the SSI version. Ease of maintenance >argues for putting SSI specific changes into the CI repository, and >deactivating them with #ifdef CONFIG_SSI, CONFIG_CFS, etc., as Aneesh >has done. > >Are there any comments on which approach would be better -- purity vs. >ease of maintenance? > >-Brian > > >--__--__-- > >Message: 3 >From: "Aneesh Kumar K.V" >To: "Brian J. Watson" >Cc: Cluster CI , > SSI discussion >Date: 26 Oct 2002 10:25:15 +0530 >Subject: [SSI] Re: [CI] Changes needed to build the CVS version of CI. > >On Sat, 2002-10-26 at 03:28, Brian J. Watson wrote: > > >>"Aneesh Kumar K.V" wrote: >> >> >>>Hi, >>> >>> Attaching below is the changes needed to build the CVS version of CI. >>>Mostly #ifdef changes. I am not sure i am using the right #ifdef . Can >>>someone verify it and do the necessary checkin. ? >>> >>> >>I'm a bit torn about the issue of SSI code existing in the CI >>repository. Ideally it shouldn't be there. Any CI file touched by SSI >>should exist in both repositories, with the SSI changes made to only the >>SSI version. >> >>Although this would be the purest approach, it also creates a bit of a >>maintenance headache for CI. Any change made to the CI version of one of >>these files must also be made to the SSI version. Ease of maintenance >>argues for putting SSI specific changes into the CI repository, and >>deactivating them with #ifdef CONFIG_SSI, CONFIG_CFS, etc., as Aneesh >>has done. >> >>Are there any comments on which approach would be better -- purity vs. >>ease of maintenance? >> >>-Brian >> >> >> >> > >I vote for ease of maintenance. The purity argument will be dropped if >we are planning to merge with Main Line Linux kernel :) ( I hope this >will happen ) > >-aneesh > > > > >--__--__-- > >Message: 4 >Subject: Re: [SSI] Queries related to SSI Linux >From: "Aneesh Kumar K.V" >To: Roopesh >Cc: ssic-linux-devel >Date: 26 Oct 2002 10:59:51 +0530 > >Hi, > > >On Fri, 2002-10-25 at 19:54, Roopesh wrote: > > >>Hi, >> >> We are evaluating the feasibilty of using SSI Linux as a >>platform >> for high availability solution. In this regard, we have the >>following >> architectural queries regarding the SSI Linux. >> >> We would appreciate if you can provide any information on these: >> >>- What is the maximum number of nodes that is supported in SSI >>without performance >> >> > >By default support 15 nodes but was tested and demonstrated with 30 >nodes at linux world. > > > >>- Is Jumbogram support possible for inter node link ? >> >> > >I did a google serach to find out what is this "Jumbogram" and found >" Using the Internet Protocol version 6 (IPv6), a jumbogram is a >transmission packet that contains a payload larger than 65,535 eight-bit >bytes (also known as octets). " > >So if it is anything related ti ipv6 then no. internode communication is >based on ipv4. I have seen data structures that are capable of carrying >only 32bits char[4]. > >But the ICS layer is highly modular and one can implement one using ipv6 >. the current implementation can be used as a reference., > > > >>- Can fibre channel be used as a lower level transport for ICS ? >> >> > >IP is the method used for communication. But as i said above it can be >extended easily. That is what i feel after looking into the code. > > > > >>- Is broadcast and multicast support available for inter node >>communication ? >> >> > >I don't think so. But i could be wrong. I am finding separate channels( >socket being opened for lower level nodes to higher level nodes in ICS >). I didn't find anything related to broadcast or multicast. > > > >>- Is it possible for one to build a custom messaging service >>within the kernel which uses ICS ? >> >> > >If you looking for a kernel based messaging. YES. Take a look at. the >latest ipvs-bind patch i send. It is much simpler and can easily be >understood. BTW a find *.svc on the /usr/src/linux/cluster should give >you more such example. > > > > > >>- Can a kernel component use low level ICS hooks directly ? >> >> >ICS exposes certain API's which can be used by any part of kernel . take >a look at /usr/src/linux/cluster/util/cluster_ksyms.c > > > >>- Does the messaging interface available in SSI, provide reliable >>sequenced delivery ? >> >> > >I guess so. If i get it correctly it uses tcp. But there were some >discussion about UDP. I am not sure about the same. > > > >>- In comparison to NSC, what are the different data integrity >>levels supported in SSI linux ? >> >> > >:) > > > >>- Is there a clustered volume manager available in SSI-linux ? >>What are its features and what stage >> of development is it in ? >> >> > >See the project page at sourceforge for more information. > > > >>- How reliable is SSI linux to address a requirement for building >>a production quality clustering solution >> using blade servers ? >> >> > >Give it a try and decide yourself. :) > > > >>- Are there performance figures available for SSI linux >>(especially max >> data rates on the inter node links) ? >> >> > >I haven't seen any mail that discuss the same. > > >>- Is the performance of SSI linearly scalable on addition of new >>nodes in the cluster ? >>- What is the plan to provide time synchronization between nodes ? >> >> > >I guess Roopesh has taken up the work ( That is what i found from the >source forge project page ). But i never heard anything from him on the >list. Is that you ? :) > > > >>- Does CFS support failover between the nodes (root & secondary)? >> >> > >David Zafman is working on this. > > > >>- We are considering a configuration where all the nodes can >>directly access the shared disk without >> being served by a master node. Which is the appropriate >>filesystem layer you suggest (CFS or GFS)? >> >> >> > >GFS > > > >>Thanks, >>Roopesh >> >> > >-aneesh > > > > >--__--__-- > >Message: 5 >From: "Sharad Tiwari" >To: >Reply-To: "Sharad Tiwari" >Date: Sat, 26 Oct 2002 18:22:40 +0530 >Subject: [SSI] problem adding new node > >hi all, > >We are trying to add a new node to the cluster ... we have SSI installed on one PC and are booting another PC from floppy which has an Etherboot image (it does not have SSI installed in it) ... > >Now when this second PC is booted it probes for a dhcp server ... and then tries to boot from network but fails saying that it could not read from boot device (which is /dev/ram0) ... > >couls someone please help me out here and give e pointers as to what could be the problem and how I should go about this ... > >thanking you in advance for the help > >Cheers!!!! >Best Wishes and Warm regads >Sharad > >Get Your Private, Free E-mail from Indiatimes at http://email.indiatimes.com > > Buy Music, Video, CD-ROM, Audio-Books and Music Accessories from http://www.planetm.co.in > >Change the way you talk. Indiatimes presents Valufon, Your PC to Phone service with clear voice at rates far less than the normal ISD rates. Go to http://www.valufon.indiatimes.com. Choose your plan. BUY NOW. > > > > >--__--__-- > >_______________________________________________ >ssic-linux-devel mailing list >ssi...@li... >https://lists.sourceforge.net/lists/listinfo/ssic-linux-devel > > >End of ssic-linux-devel Digest > > ------------------------------------------------------- This SF.net email is sponsored by: ApacheCon, November 18-21 in Las Vegas (supported by COMDEX), the only Apache event to be fully supported by the ASF. http://www.apachecon.com _______________________________________________ ssic-linux-devel mailing list ssi...@li... https://lists.sourceforge.net/lists/listinfo/ssic-linux-devel Get Your Private, Free E-mail from Indiatimes at http://email.indiatimes.com Buy Music, Video, CD-ROM, Audio-Books and Music Accessories from http://www.planetm.co.in Change the way you talk. Indiatimes presents "Valufon", Your PC to Phone service with clear voice at rates far less than the normal ISD rates. Go to http://www.valufon.indiatimes.com. Choose your plan. BUY NOW. |