October 16, 2011 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >
October 16, 2011 [00:00:07] <tsoome> both the switch and the array should see the hba wwn[00:00:55] <tsoome> if cfgadm -al does not list array at all, it hints either HBA/FC stack issue in solaris, or zoning issue in switch[00:01:37] <fec> should that list the array or the pci devices?[00:01:45] <fec> it does show the pcie devices....[00:01:52] <tsoome> cfgadm should list both[00:02:09] <fec> c2 fc connected unconfigured unknown[00:02:09] <fec> c3 fc-fabric connected unconfigured unknown[00:02:09] <fec> pcie160 fibre/hp connected configured ok[00:02:09] <tsoome> c1 fc-private connected configured unknown[00:02:09] <tsoome> c1::21000014c3ca937a disk connected configured unknown[00:02:33] <fec> pcie160 is the qlc afaik[00:02:41] <tsoome> ok, so your FC stack does see HBA and it does know its connected to fabric[00:03:00] <fec> what does c1/2 etc.. mean?[00:03:19] <tsoome> controller[00:03:51] <tsoome> ah, well, your system does see c2 and c3, but only c3 is fabric[00:04:44] <tsoome> can you pastebin entire output from cfgadm -al[00:05:08] <fec> So pcie160 is FC attached to a jbod shelf, pcie192 is to a mds to an emc[00:06:05] <fec> Ap_Id Type Receptacle Occupant Condition[00:06:05] <fec> c1 scsi-bus connected configured unknown[00:06:06] <fec> c1::dsk/c1t0d0 disk connected configured unknown[00:06:06] <fec> c1::scsi/direct/c1t0d0 disk connected configured unknown[00:06:06] <fec> c2 fc connected unconfigured unknown[00:06:06] <fec> c3 fc-fabric connected unconfigured unknown[00:06:06] <fec> pcie160 fibre/hp connected configured ok[00:06:07] <fec> pcie192 fibre/hp connected configured ok[00:06:24] <fec> all the rest are pcie### unkown/empty/unconfigured/unkown[00:06:39] <tsoome> do cfgadm -c configure c2 and cfgadm -c configure c3[00:07:22] <fec> c3 reports configured now, c2 still unconfigured[00:07:41] <fec> I think there is a cabling issue with the C2 as hbainfo shows disconnected[00:07:43] <tsoome> any disks from cfgadm -al ?[00:08:22] <tsoome> yep, that may be true, because on connection you should see fc-fabric or fc-private, not just fc[00:08:27] <fec> I have no lun's assigned as I cant see the hba on the emc because it has not done a perli (i am assuming)[00:08:42] <tsoome> aye, that makes sense[00:08:46] <fec> is there a way to force a bus scan?[00:10:13] <tsoome> luxadm -e forcelip /dev/cfg/c2 or c3[00:11:04] <tsoome> that will force login into fabric[00:11:57] <tsoome> anyhow, even fcinfo hba-port tells if you have link up or not[00:12:06] <tsoome> if its not up, its cabling issue[00:12:15] <tsoome> or switch port is turned off[00:13:28] <fec> yea I think the hba to DAS is cabled wrong (no port type/state) , but the HBA to FCSW is good (n-port/online)[00:14:19] <fec> or on the DAS I need to force it to be an N-port[00:14:46] <fec> I dont know point to point fc at all never played with it[00:15:38] <fec> still dont see it on the EMC side[00:15:50] <tsoome> at solaris side HBA picks normally up the correct mode, if the switch port is on autoconfigure.[00:16:39] <tsoome> at storage side you may set port mode; cant tell about EMC, HDS midrange you need to set manually to point-to-point (or loop on direct connection)[00:22:45] *** Z` has quit IRC[00:25:07] <fec> still has not showed up to the emc[00:25:50] <fec> hba - mds - emc, no zoning in the switch[00:29:06] <tsoome> does switch see both hba and emc wwn?[00:30:36] <fec> luxadm -e port[00:30:36] <fec> /devices/pci@0,0/pci15ad,7a0@15/pci1077,117@0/fp@0,0:devctl NOT CONNECTED[00:30:36] <fec> /devices/pci@0,0/pci15ad,7a0@16/pci1077,117@0/fp@0,0:devctl NOT CONNECTED[00:31:49] <fec> this is hba from switch with successful flogi login[00:31:49] <fec> fc1/12 11 0x470100 21:00:00:e0:8b:1f:de:84 20:00:00:e0:8b:1f:de:84[00:31:50] <fec> this is emc from switch with successful flogi login[00:31:51] <fec> fc1/4 11 0x4700ef 50:06:01:69:3a:60:13:8e 50:06:01:60:ba:60:13:8e[00:32:44] *** arobase has quit IRC[00:32:58] <fec> # fcinfo hba-port | grep 84[00:32:58] <fec> HBA Port WWN: 210000e08b1fde84[00:32:58] <fec> Node WWN: 200000e08b1fde84[00:33:14] <fec> fcinfo shows it state=online[00:34:14] <tsoome> so, create zone in switch[00:34:46] <tsoome> switch does see both logins, but devices wont see each other[00:51:18] *** fisted has quit IRC[01:19:16] <fec> I manually added the wwpn/nn to the emc and presented a disk and it works[01:19:38] <fec> not sure why the wwn's were never displaying to the emc[01:20:02] <tsoome> some setting in switch[01:20:24] <fec> other servers work and ports config'd same way[01:20:35] <fec> no zoning on switch, just masking on lun's (this is a lab)[01:20:55] <tsoome> uhm, thats not too wise:)[01:21:17] <fec> oh boy do i know =)[01:21:22] <tsoome> :)[01:21:55] <fec> back in the day i worked at a place with like 40 fabrics[01:22:13] <tsoome> can be just some wierd glitch in switch firmware anyhow[01:22:14] <fec> the eva admins hated zoning so had their switches without zoning, mass fabric failures are fun[01:23:32] <tsoome> that sounds like job for an real men;)[01:23:38] <tsoome> with balls;)[01:23:43] <fec> trying to build a solaris machine to replace this horrible emc in a lab, not so familiar with the disk shelves they provided so put this solaris machine on a vsan that the emc is on to see if I could atleast get the solaris portion working[01:24:03] <fec> next need to figure out how to make these DAS FL disks work, then how to make hba targets[01:24:30] <tsoome> which solaris it is?[01:24:42] <fec> nexenta[01:24:53] <fec> so off opensolaris b314 i think[01:25:28] <tsoome> depending on hba, you need to switch initiator driver to target on ports you need, rest is usual comstar setup.[01:25:50] <tsoome> need to have quite recent fabric tho[01:25:55] <fec> yea was reading about switching to QLT, I am assuming that is the driver needed to set hba in target mode[01:26:45] <tsoome> emlex is easier - just line in emlxl.conf, qlogic needs driver swap[01:26:56] <tsoome> emulex*[01:27:27] <fec> Do you have a target fc setup on solaris?[01:27:36] <tsoome> I did the test with emulex[01:27:52] <fec> I am wondering if nexenta is what I should be using, or opensolaris or openindiana[01:28:10] <tsoome> once you have target port, rest is just like with iscsi or fcoe[01:28:45] <fec> Eventually plan to setup some fcoe targets as well, just not sure what distribution I should use as a base[01:28:47] <tsoome> just instead of iqn you have wwn to identify your target port[01:29:31] <tsoome> I couldnt test it too much since I only have brocade silkworm 3800 as switch and its an shwstopper[01:29:56] <tsoome> but direct connection was working just fine[01:30:39] <tsoome> I did test on s11 express tho (had T5240 servers to play with)[01:31:22] *** kimc has quit IRC[01:32:37] <tsoome> fcoe needs dedicated nic and jumbo frames.[01:35:55] *** TomJ has quit IRC[01:46:25] *** stux|RC-only has quit IRC[01:49:08] *** stux|RC-only has joined #opensolaris[01:51:21] *** [JT] has joined #opensolaris[01:52:11] *** ewdafa has quit IRC[02:20:27] *** [JT] has left #opensolaris[02:33:36] *** AxeZ has quit IRC[02:45:38] *** yippi has joined #opensolaris[02:46:20] *** jimmy1980 has quit IRC[02:55:32] *** jbk has quit IRC[03:01:28] *** jbk has joined #opensolaris[03:11:47] *** jbk has quit IRC[03:31:37] *** Mdx4 has joined #opensolaris[04:20:36] *** Mdx4 has quit IRC[04:23:28] *** Tpenta has quit IRC[04:24:20] *** Tpenta has joined #opensolaris[04:46:40] *** yippi has quit IRC[04:47:58] *** yippi has joined #opensolaris[05:08:27] *** Tpenta has quit IRC[05:08:40] *** yippi has quit IRC[05:10:10] *** Tpenta has joined #opensolaris[05:20:16] *** yippi has joined #opensolaris[05:27:17] *** jayne has quit IRC[05:33:01] *** jayne has joined #opensolaris[05:39:01] *** jbk has joined #opensolaris[05:50:13] *** jbk has quit IRC[06:05:03] *** comay has quit IRC[06:08:16] *** ChanServ sets mode: +o Tpenta[06:08:53] *** smrt has quit IRC[06:09:08] *** smrt has joined #opensolaris[06:13:54] *** comay has joined #opensolaris[06:17:35] *** ChanServ sets mode: +o comay[06:25:36] *** dajhorn has quit IRC[06:44:54] *** jbk has joined #opensolaris[06:49:37] *** jbk has quit IRC[06:59:08] *** gwr has quit IRC[07:35:04] <fec> Is there a way to list disks other than 'format' ? 'format' takes forever as I think I have a scsi device with some errors[07:44:11] <Triskelios> fec: geez, don't spam your question everywhere[07:51:33] <fec> Its late Trisk not many are awake sorry.[08:12:00] *** JoergB has quit IRC[08:12:21] *** JoergB has joined #opensolaris[08:33:13] *** uw__ has joined #opensolaris[08:49:20] *** hajma has joined #opensolaris[09:04:54] *** jbk has joined #opensolaris[09:08:20] *** dduvall has quit IRC[09:09:20] *** dduvall has joined #opensolaris[09:26:24] *** ingenthr has quit IRC[09:31:40] *** yippi has quit IRC[09:35:54] <tsoome> iostat -En will list as well[09:40:50] *** mceith has quit IRC[09:42:08] *** mceith has joined #opensolaris[09:43:01] *** kdavy has quit IRC[09:59:10] *** kdavy has joined #opensolaris[10:03:47] *** kdavy has quit IRC[10:05:14] *** kdavy has joined #opensolaris[10:08:21] *** alanc has quit IRC[10:10:58] *** Meths has quit IRC[10:11:08] *** Meths has joined #opensolaris[10:14:03] *** NorrinRadd has joined #opensolaris[10:17:04] *** Norrin has quit IRC[10:19:00] *** Norrin has joined #opensolaris[10:19:05] *** NorrinRadd has quit IRC[10:57:49] *** JoergB has quit IRC[10:58:04] *** JoergB has joined #opensolaris[11:04:35] *** Mdx4 has joined #opensolaris[11:11:05] *** ginabc has joined #opensolaris[11:15:31] *** ginabc has quit IRC[11:24:04] *** Mdx4 has quit IRC[11:59:00] *** kdavy has quit IRC[12:05:08] *** hguy has quit IRC[12:05:57] *** hguy has joined #opensolaris[12:08:20] *** comay has quit IRC[12:09:32] *** ewdafa has joined #opensolaris[12:10:18] *** herzen has joined #opensolaris[12:10:26] *** CVLTCMK0 has quit IRC[12:18:40] *** RoyK has joined #opensolaris[12:27:27] *** CVLTCMK0 has joined #opensolaris[12:51:38] *** CVLTCMK0 has quit IRC[13:01:48] *** CVLTCMK0 has joined #opensolaris[13:11:59] *** RoyK has quit IRC[13:13:12] *** jimmy1980 has joined #opensolaris[13:14:41] *** RoyK has joined #opensolaris[13:17:17] *** wdp_ has joined #opensolaris[13:38:50] *** fec has quit IRC[13:55:04] *** fisted has joined #opensolaris[14:10:28] *** troydm has joined #opensolaris[14:10:44] <troydm> i have upgraded to b134 from b111 OpenSolaris and now xvnc isn't starting twm[14:10:52] <troydm> how can i fix that?[14:14:50] *** TomJ has joined #opensolaris[14:14:50] *** TomJ has joined #opensolaris[14:18:39] *** TomJ has quit IRC[14:21:54] *** TomJ has joined #opensolaris[14:21:54] *** TomJ has joined #opensolaris[14:23:05] *** TomJ has quit IRC[14:27:26] <lblume> troydm: By upgrading to a non-dead operating system?[14:28:10] <troydm> lblume: what do you mean by that?[14:28:50] <lblume> OpenSolaris is dead. What is the point in running it?[14:30:20] <troydm> lblume: well. as long as it works and is stable it's not dead[14:31:59] <lblume> Duh, as long as its publisher said it's dead, then it's dead. If you have issues with it, you're pretty much on your own.[14:38:36] *** jimmy1980 has quit IRC[14:41:16] <CosmicDJ> so this room is full of necrophiles? ;)[14:41:51] <lblume> CosmicDJ: Kinda :-D[14:42:27] <lblume> Also necromancers who manage to reanimate the thing and turn it into something less dead :-)[14:44:21] *** jimmy1980 has joined #opensolaris[14:54:52] *** fisted_ has joined #opensolaris[14:57:33] *** fisted has quit IRC[14:59:12] <troydm> no one uses xvnc-inetd on b134?[15:00:41] <tsoome> sorry, no b134 installs left...[15:02:13] <troydm> ok it seems that in order to make it fix i need to add .vnc/xstartup[15:02:27] <troydm> however i don't know which folder to put it in[15:02:52] <troydm> /.vnc/xstartup ?[15:02:59] <troydm> or /root/.vnc/xstartup ?[15:10:55] <tomww> cd[15:10:58] <lblume> b134, like any unfinished system that was never released, and is not updated, is full of issues, known or not, that will never get fixed, and at this point, are of no interest to most people, unless they want to upgrade it to something current.[15:12:09] <eklof> I use Solaris 11 EA, seems quite solid atm.[15:29:47] *** fec has joined #opensolaris[15:43:05] *** fec has quit IRC[15:43:28] *** fOB has joined #opensolaris[16:00:55] *** TomJ has joined #opensolaris[16:00:57] *** TomJ has quit IRC[16:00:57] *** TomJ has joined #opensolaris[16:11:37] *** RoyK has quit IRC[16:17:11] *** jimmy1980 has quit IRC[16:21:55] *** mavhc has quit IRC[16:22:16] *** hajma has quit IRC[16:23:11] *** mavhc has joined #opensolaris[16:23:41] *** jimmy1980 has joined #opensolaris[16:24:12] *** RoyK has joined #opensolaris[16:24:36] *** mavhc has joined #opensolaris[16:25:25] *** TomJ has quit IRC[16:40:58] *** RoyK has quit IRC[16:43:44] *** RoyK has joined #opensolaris[16:48:00] *** gray-- has joined #opensolaris[17:03:33] *** AxeZ has joined #opensolaris[17:17:21] *** rorx has quit IRC[17:38:17] *** echobinary has quit IRC[17:46:29] *** plat- has quit IRC[17:57:37] *** plat- has joined #opensolaris[18:12:46] *** alanc has joined #opensolaris[18:12:46] *** ChanServ sets mode: +o alanc[18:18:44] *** fec has joined #opensolaris[18:38:01] *** gray-- has quit IRC[18:57:23] *** kloeri has quit IRC[18:57:45] <jelmd> v440: is there a way to determine from the CLI, whether a ce (cassini?) based NIC has Cu or Fiber ports?[19:11:40] *** echobinary has joined #opensolaris[19:18:24] *** tsoome has quit IRC[19:21:42] *** tsoome has joined #opensolaris[19:27:47] *** hajma has joined #opensolaris[19:41:13] *** kloeri has joined #opensolaris[19:42:17] *** ingenthr has joined #opensolaris[19:46:48] *** thuttu77 has quit IRC[19:48:04] *** kloeri has quit IRC[19:50:21] *** thuttu77 has joined #opensolaris[20:06:54] *** Mdx4 has joined #opensolaris[20:07:57] *** gray-- has joined #opensolaris[20:08:08] *** gray-- has quit IRC[20:21:32] *** herzen has left #opensolaris[20:42:26] *** fec has quit IRC[20:44:37] *** fec has joined #opensolaris[20:57:19] *** Seeder has joined #opensolaris[20:57:34] <fec> When I type the 'format' command it never returns just kinda times out, how should I start troubleshooting this?[20:58:57] <fec> There is a FL-AL device attached to a HBA and this is what I think is causing the issue, do I have to set the HBA to anything special?[20:59:13] *** stux|RC-only has quit IRC[21:00:40] *** stux|RC-only has joined #opensolaris[21:02:20] *** fOB has quit IRC[21:11:55] *** kloeri_ has joined #opensolaris[21:30:59] *** kloeri_ is now known as kloeri[21:37:50] *** rorx has joined #opensolaris[22:06:00] *** ivenkys has left #opensolaris[22:13:06] <tsoome> usually no; format reads inquiry data for vendor and label, timeout means it has hard time to get reads done.[22:17:58] *** jimmy1980 has quit IRC[22:20:28] <fec> What log files would contain errors I can debug?[22:23:03] <wdp_> any.[22:24:08] *** jimmy1980 has joined #opensolaris[22:24:20] <sommerfeld> iostat -En[22:38:34] <tsoome> /var/adm/messages[22:51:12] <CIA-56> SFE jurikm: SFEfreeglut.spec: add IPS package name[22:52:17] <CIA-56> SFE jurikm: SFEyasm.spec: add IPS package name[22:53:33] <CIA-56> SFE jurikm: encumbered/SFEgpac.spec: add IPS package name[22:58:11] <CIA-56> SFE jurikm: SFElibmp4v2.spec: add IPS package name base-specs/libmp4v2.spec: fix sun studio build[22:59:30] *** wdp_ has quit IRC[23:05:23] <CIA-56> SFE jurikm: encumbered/SFEfaad2.spec: add IPS package name, fix build[23:22:45] <CIA-56> SFE jurikm: encumbered/SFElibx264.spec: add IPS package name, keep SFEgpac as mandatory base-specs/libx264.spec: fix multicore build[23:27:28] *** Silikon has joined #opensolaris[23:29:21] <CIA-56> SFE jurikm: SFEgiflib.spec: add IPS package name[23:32:31] *** troydm has quit IRC[23:54:10] *** Meths has quit IRC