<html><body><p><font size="2">We have not tested such a procedure. The only route that we have done is a complete mmdelnode/mmaddnode scenario. This would mean an mmdeldisk. It would be more time consuming since data has to move. </font><br><font size="2">Operating in a mixed architecture environment is not a problem. We have tested and support that. </font><br><font size="2"><br><br>Steve Duersch<br>Spectrum Scale<br>845-433-7902<br>IBM Poughkeepsie, New York<br><br><br></font><br><br><tt><font size="2"><br>> <br>> Message: 1<br>> Date: Wed, 7 Jun 2017 14:04:28 +0000<br>> From: "Sundermann, Jan Erik (SCC)" <jan.sundermann@kit.edu><br>> To: "gpfsug-discuss@spectrumscale.org"<br>> <gpfsug-discuss@spectrumscale.org><br>> Subject: [gpfsug-discuss] Upgrade with architecture change<br>> Message-ID: <E0E3879D-A6EB-40FD-B1E7-F9E0559E3E98@kit.edu><br>> Content-Type: text/plain; charset="utf-8"<br>> <br>> Hi,<br>> <br>> we are operating a small Spectrum Scale cluster with about 100 <br>> clients and 6 NSD servers. The cluster is FPO-enabled. For <br>> historical reasons the NSD servers are running on ppc64 while the <br>> clients are a mixture of ppc64le and x86_64 machines. Most machines <br>> are running Red Hat Enterprise Linux 7 but we also have few machines<br>> running AIX.<br>> <br>> At the moment we have installed Spectrum Scale version 4.1.1 but <br>> would like to do an upgrade to 4.2.3. In the course of the upgrade <br>> we would like to change the architecture of all NSD servers and <br>> reinstall them with ppc64le instead of ppc64. <br>> <br>> From what I?ve learned so far it should be possible to upgrade <br>> directly from 4.1.1 to 4.2.3. Before doing the upgrade we would like<br>> to ask for some advice on the best strategy. <br>> <br>> For the NSD servers, one by one, we are thinking about doing the following:<br>> <br>> 1) Disable auto recovery<br>> 2) Unmount GPFS file system<br>> 3) Suspend disks<br>> 4) Shutdown gpfs<br>> 5) Reboot and reinstall with changed architecture ppc64le<br>> 6) Install gpfs 4.2.3<br>> 7) Recover cluster config using mmsdrrestore<br>> 8) Resume and start disks<br>> 9) Reenable auto recovery<br>> <br>> Can GPFS handle the change of the NSD server?s architecture and <br>> would it be fine to operate a mixture of different architectures for<br>> the NSD servers?<br>> <br>> <br>> Thanks,<br>> Jan Erik<br>> <br>> <br></font></tt><BR>
</body></html>