similar to: Inconsistent behavior between x86_64 and ppc64 when creating guests with NUMA node placement

Solved. For the archives: Noticed in xl info output that only 1 core was recognised, even though nr_cpus should show 16 on this box. Tried rebooting and selecting the old .34 kernel, console printed "smpboot: do_boot_cpu failed(-1) to wakeup CPU#1" and CentOS failed to boot. Powered down the box fully and started back up with the latest .44 kernel and everything is working fine

About rationale, usage and (some small bits of) API. Signed-off-by: Dario Faggioli <dario.faggioli@citrix.com> Acked-by: Ian Campbell <ian.campbell@citrix.com> --- Changes from v7: * avoid referring to 4.2 release as "upcoming". * libxl placement disabling key explicitly mentioned. * Limit of max 16 NUMA nodes explicitly mentioned. Changes from v6: * text updated to

I just did a yum update on a CentOS 7 / Xen 4.6 server which took me from kernel-3.18.34-20.el7.x86_64 -> kernel-3.18.44-20.el7.x86_64 After rebooting, the following notice is printed immediately upon xl create'ing a domain: libxl: notice: libxl_numa.c:499:libxl__get_numa_candidate: NUMA placement failed, performance might be affected Indeed performance is significantly degraded. This

Hi, this patch introduces a new config file option (numanodes=[x]) to specify a list of valid NUMA nodes for guests. This will extend (but not replace) the recently introduced automatic placement. If several nodes are given, the current algorithm will choose one of them. If none of the given nodes has enough memory, this will fall back to the automatic placement. Signed-off-by: Andre

Help I can produce the hexidecimal equivalent of a decimal number but I am having a hard time reversing the operation. I'm good for hex representations to 159 and am close to extending to 2559. The archives are not clear on the existence of a function for this task. Is there one? Here is what I have got so far: #Good for hex values to "9F" as.decmode<-function(as.character(x)){

> a <- 0 ; for(i in (1:200000000)) a <- a + 1/i > b <- 0 ; for(i in (200000000:1)) b <- b + 1/i > c <- sum(1/(1:200000000)) > d <- sum(1/(200000000:1)) > order(c(a,b,c,d)) [1] 1 2 4 3 > b<c [1] TRUE > c==d [1] FALSE I'd expected b being the largest, since we sum up the smallest numbers first. Instead, c is the largest, which is

Hi, Our lustre environment is: 2.6.9-55.0.9.EL_lustre.1.6.3smp I would like to change recovery timeout from default value 250s to something longer I tried example from manual: set_timeout <secs> Sets the timeout (obd_timeout) for a server to wait before failing recovery. We performed that experiment on our test lustre installation with one OST. storage02 is our OSS [root at

Hi, Lustre manual 1.6 v18 says that that in production lustre debug level should be set to fairly low. Manual also says that I can verify that level by running following commands: # sysctl portals.debug This gives ne following error error: ''portals.debug'' is an unknown key cat /proc/sys/lnet/debug gives output: ioctl neterror warning error emerg ha config console cat

In 1.6.0, when creating a MDT, you could specify multiple --mgsnode options and it would failover between them. 1.6.3 only seems to take the last one and --mgsnode=192.168.1.252 at o2ib:192.168.1.253 at o2ib doesn''t seem to failover to the other node. Any ideas how to get around this? Robert Robert LeBlanc College of Life Sciences Computer Support Brigham Young University leblanc at

I am updateding my LDAP presentation (ftp://kalamazoolinux.org/pub/pdf/ldapv3.pdf) to include a section on the Samba 2.2.x LDAP backend. But I can't find any definititive documentation on the "time" attributes: pwdLastChange, pwdMustChange. Either how the time is encoded or exactly what they mean. I've seen post saying they are hexidecimal (although they don't look it)

Hi Folks, We can see these massages on our MDS Feb 12 12:46:08 mds01 kernel: LDISKFS-fs warning (device dm-0): empty_dir: bad directory (dir #31452569) - no `.'' or `..'' Feb 12 12:46:08 mds01 kernel: LDISKFS-fs warning (device dm-0): ldiskfs_rmdir: empty directory has too many links (3) It seem to indicate that we have bad(corrupted) directory. Do you have any idea how to

From: Dave Airlie <airlied at redhat.com> When a buffer is evicted for memory pressure or TTM evict all, the placement is set to the eviction domain, this means the buffer never gets revalidated on the next exec to the correct domain. I think this should be fine to use the initial domain from the object creation, as least with VM_BIND this won't change after init so this should be the

One of the greatest things about rails is that it is so standards-compliant, no other framework that I have seen have complied to the HTTP standard (think REST) in such a degree that Rails does. Kudos to you all for that. I think we (Rails community) should follow the line of standards compliance and also take it to the binary prefixes [1], i.e. kilobytes, megabytes, etc. For more than half a

Hi everybody, i'm getting mad configuring samba to join an ADS, resolve domain users and groups and set ACLs via windows explorer on a share mounted with POSIX ACL and extended attributes. At the point where i am, i've managed to get Samba join correctly the domain with idmap_rid backend working fine. I can correctly set (add, remove, modify) file acls and extended attributes

Hello, We have running lustre 1.8.1 and have met "No space lest on device" error when uploading 500 Gb small files (less then 100 Kb each). The problem seems to depends on the number of files. If we remove one file, we can create one new file, even with Gb size; but if we haven''t remove something we can''t create even very little file, as an example using touch

core and the simple menu do not interpret the IPAPPEND/SYSAPPEND directives in the same way. Which is the proper way? Either way, this should be clarified in the documentation. com32/elflink/ldlinux/readconfig.c: } else if ((ep = looking_at(p, "ipappend")) || (ep = looking_at(p, "sysappend"))) { uint32_t s = strtoul(skipspace(ep), NULL,

Hi, Our environment is: 2.6.9-55.0.9.EL_lustre.1.6.3smp I am getting following errors from two OSS''s ... Nov 7 10:39:51 storage09.beowulf.cluster kernel: LustreError: 23045:0:(client.c:519:ptlrpc_import_delay_req()) @@@ IMP_INVALID req at 00000100b410be00 x4190687/t0 o101->MGS at MGC10.143.245.201@tcp_0:26 lens 232/240 ref 1 fl Rpc:/0/0 rc 0/0 Nov 7 10:39:51

I would like to call an image to be displayed based on its title. It is easy enough for me to do this like this: <img src="/images/suggested/<%= suggested.title %>" height="100" width="100" /> But I want to stick within the ruby code and I''m trying this: <%= image_tag("/images/suggested/<%= suggested.title %>", size=>

<link rel="shortcut icon" href="/public/favicon.ico" > This causes a routing error, and the icon doesn''t load. Should I add something to my routes.rb file, or should I change the href in the above link? -- Posted via http://www.ruby-forum.com/.

Using Thomas Baier's DCOM server I can do the following (in Windows 2000): I create a slider in Excel, and the sliders On_Change event triggers a call to R. This call redraws an R graph, and so I have animated R graphics controlled by Excel's slider. The problem is that Excel needs focus so i can grab the slider, and therefore it partially covers R's graph window. Is there a way to