similar to: RDMA inline threshold?

Forgot to mention, sometimes I have to do force start other volumes as well, its hard to determine which brick process is locked up from the logs. Status of volume: rhev_vms_primary Gluster process TCP Port RDMA Port Online Pid ------------------------------------------------------------------------------ Brick spidey.ib.runlevelone.lan:/gluster/brick/rhev_vms_primary

Stefan, Sounds like a brick process is not running. I have notice some strangeness in my lab when using RDMA, I often have to forcibly restart the brick process, often as in every single time I do a major operation, add a new volume, remove a volume, stop a volume, etc. gluster volume status <vol> Does any of the self heal daemons show N/A? If that's the case, try forcing a restart on

Dear Dan, thanks for the quick reply! I actually tried restarting all processes (and even rebooting all servers), but the error persists. I can also confirm that all birck processes are running. My volume is a distrubute-only volume (not dispersed, no sharding). I also tried mounting with use_readdirp=no, because the error seems to be connected to readdirp, but this option does not change

Hi Stefan, There is a known issue with gluster 3.12.x builds (see [1]) so you may be running into this. Please take a look at [1] and try out the workaround provided in the comments. Regards, Nithya [1] https://bugzilla.redhat.com/show_bug.cgi?id=1517260 On 9 March 2018 at 13:37, Stefan Solbrig <stefan.solbrig at ur.de> wrote: > Dear all, > > I have a problem with df after

Dear all, I have a problem with df after I upgraded from 3.12.4 to 3.12.6 All four bricks are shown als online, and all bricks are being used. gluster v stats shows the correct sizes for all devices. However, df does not show the correct glusterfs volume size. It seems to me that it "forgets" one brick. Although all bricks are used when I'm writing files. best wishes, Stefan

Dear all, I have problem with glusterfs 3.12.4 mkdir -p fails with "no data available" when umask is 0022, but works when umask is 0002. Also recursive copy (cp -R or cp -r) fails with "no data available", independly of the umask. See below for an example to reproduce the error. I already tried to change transport from rdma to tcp. (Changing the transport works, but

Hello gluster users and professionals, We are running gluster 3.10.10 distributed volume (9 nodes) using RDMA transport. From time to time applications crash with I/O errors (can't access file) and in the client logs we can see messages like: [2018-05-04 10:00:43.467490] W [MSGID: 114031] [client-rpc-fops.c:2640:client3_3_readdirp_cbk] 0-gv0-client-2: remote operation failed [Transport

Dear all, I encountered the same issue, I saw that this is fixed in 3.12.7 but I cannot find this release in the main repo (centos storage SIG), only in the test one. What is the expectation to see this release available into the main repo? Greetings, ??? Paolo Il 09/03/2018 10:41, Stefan Solbrig ha scritto: > Dear Nithya, > > Thank you so much. ?This fixed the problem

Dear all, I identified the source of the problem: if i set "server.root-squash on", then the problem is 100% reproducible, with "server.root-squash off", the problem vanishes. This is true for glusterfs 3.12.3, 3.12.4 and 3.12.6 (haven't tested other versions) best wishes, Stefan -- Dr. Stefan Solbrig Universit?t Regensburg, Fakult?t f?r Physik, 93040 Regensburg,

Hi Alex, Thank you for the quick reply! Yes, I'm aware that using ?plain? hardware with replication is more what GlusterFS is for. I cannot talk about prices where in detail, but for me, it evens more or less out. Moreover, I have more SAN that I'd rather re-use (because of Lustre) than buy new hardware. I'll test more to understand what precisely "replace-brick"

Dear all, I'm rather new to glusterfs but have some experience running lager lustre and beegfs installations. These filesystems provide active/active failover. Now, I discovered that I can also do this in glusterfs, although I didn't find detailed documentation about it. (I'm using glusterfs 3.10.8) So my question is: can I really use glusterfs to do failover in the way described

Hi, The method depends a bit if you use a distributed-only system (like me) or a replicated setting. I'm using a distributed-only setting (many bricks on different servers, but no replication). All my servers boot via network, i.e., on a start, it's like a new host. To rescue the old bricks, just set up a new server this the same OS, the same IP and and the same hostname (!very

Hi Stefan, I think what you propose will work, though you should test it thoroughly. I think more generally, "the GlusterFS way" would be to use 2-way replication instead of a distributed volume; then you can lose one of your servers without outage. And re-synchronize when it comes back up. Chances are if you weren't using the SAN volumes; you could have purchased two servers

Hi, I''ve been investigating various caching methods provided by Rail. I first looked at the hash_cache module and function. In testing it, I noticed it wasn''t actually caching anything. Then looking at the source code, I noticed that it attempted to hold its cache in a class variable - which won''t actually be saved from page request to page because of the way that rails

Hi, I''m a newbie even though I''ve play with rails for a few months now. I would like to save several activerecord objects (not in the database) accross several screens. What is the prefered rails way to do this? Should copy all of their data to @session or use the member variables and put them into hidden fields? I am trying to move from "whatever kludge works" to

On Tue, Feb 04, 2014 at 04:01:32PM +0100, Pino Toscano wrote: > +static char * > +extract_optionally_quoted (const char *value) > +{ > + size_t value_len = strlen (value); > + > + if (value_len >= 2 && > + ((value[0] == '\'' && value[value_len - 1] == '\'') || > + (value[0] == '"' &&

Hi all, with the option "cluster.min-free-disk" set, glusterfs avoids placing files bricks that are "too full". I'd like to understand when the free space on the bricks is calculated. It seems to me that this does not happen for every write call (naturally) but at some interval or that some other event triggers this. i.e, if I write two files quickly (that together

Hi all, I know there are many tutorials on how to replace a gluster host that has become unusable. But they all seem to assume that the bricks of the respective host are gone, too. My problem is different and (I hope) more easily solved: the disk with the host?s root file system died and cannot be recovered. However, all of its bricks are on separate disks and completely undamaged. I?m seeking

Oh, Boy. I am in trouble now.... I just installed Centos on a USB drive on my corp notebook. To not TOUCH my corp drive. I spent time with the drive partitioner to make sure that nothing was done to the internal hard drive... Well I missed something and I overwrote the encrypted bootloader on the hard drive. Now what? Can I rescue things? I am leaving for my flight in a couple hours to

I was delighted I with how fast (faster than a tickled toddler giggles) r2u scarfed packages I. up my Pop!_OS (Ubuntu in party attire tuned for System76). But, despite 40 years as my own sys admin (classic fool for a client?), I bolluxed up, ending up with an install chain that was inferior to compiling from source, despite having to occasionally track down system dependencies. Absolutely no