Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2016-09-14

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:17 Guest35265 morning
00:17 misc nigelb !
00:17 nigelb joined #gluster-dev
00:18 nigelb misc: It's just about 6 am. Not *that* early :)
00:18 misc nigelb: I was waiting for you since the freenode split
00:18 misc and now
00:18 misc I forgot why :(
00:19 nigelb something to do with cgit?
00:19 misc nope
00:19 misc well, maybe
00:19 nigelb or cage?
00:19 misc I did a test run of the playbook on a RHEL5, and except a few bug, they should work
00:20 nigelb \o/
00:20 misc (for base)
00:20 misc and found why ansible was blocked, due to a i/o error on a one server, likely due to the ilo/idrac/whatever virtual cd
00:21 misc (but it was blocked only for regular run, so that explain also why some stuff didn't deploy as I tought they should)
00:21 misc but well, gonna sleep on it
00:23 nigelb if I'm already awake
00:23 nigelb you should sleep!
00:47 JoeJulian Is there any way to force a file to move to a different brick, similar to what rebalance does, but moves to a brick that's not the one that matches the dht mask?
01:48 ilbot3 joined #gluster-dev
01:48 Topic for #gluster-dev is now Gluster Development Channel - http://gluster.org | For general chat go to #gluster | Patches - http://review.gluster.org/ | Channel Logs - https://botbot.me/freenode/gluster-dev/ & http://irclog.perlgeek.de/gluster-dev/
02:15 magrawal joined #gluster-dev
02:59 nbalacha joined #gluster-dev
03:24 kdhananjay joined #gluster-dev
03:34 gem joined #gluster-dev
03:41 itisravi joined #gluster-dev
03:50 atinm joined #gluster-dev
04:01 aravindavk joined #gluster-dev
04:10 karthik_ joined #gluster-dev
04:14 mchangir joined #gluster-dev
04:15 kotreshhr joined #gluster-dev
04:29 itisravi kkeithley: just curious, who is  'World+Dog'  in the commit message of http://review.gluster.org/14838 ?
04:31 * itisravi was going through libglusterfs/src/syscall.c and saw this reference.
04:33 ashiq joined #gluster-dev
04:40 spalai joined #gluster-dev
04:56 ramky joined #gluster-dev
04:56 rafi joined #gluster-dev
04:58 spalai joined #gluster-dev
05:00 ndarshan joined #gluster-dev
05:03 Saravanakmr joined #gluster-dev
05:04 skoduri joined #gluster-dev
05:05 kdhananjay joined #gluster-dev
05:13 kshlm joined #gluster-dev
05:13 aspandey joined #gluster-dev
05:13 karthik_ joined #gluster-dev
05:16 JoeJulian itisravi: http://www.urbandictionary.com/d​efine.php?term=world%20%2B%20dog
05:17 ndevos itisravi: hah, thats what I was thinking too and had to duckduckgo it
05:18 itisravi JoeJulian: Thanks, today I Learned. :)
05:18 JoeJulian Me too
05:18 kkeithley Everyone and his brother!
05:19 itisravi ndevos: I googled "world+dog" and nothing turned up :)
05:19 JoeJulian Me too, then I tried "world plus dog" and found it right off.
05:19 ndevos itisravi: it been a while, maybe I went to urbandictionary too
05:19 itisravi yup
05:19 itisravi what JoeJulian said.
05:20 aravindavk joined #gluster-dev
05:20 kkeithley bob's your uncle
05:21 kkeithley lots of fun stuff in English
05:21 JoeJulian I actually do have an uncle Bob.
05:24 k4n0 joined #gluster-dev
05:31 aravindavk ndevos: please review this patch, spec file change wrt events feature http://review.gluster.org/#/c/15486
05:37 kdhananjay joined #gluster-dev
05:41 jiffin joined #gluster-dev
05:46 ppai joined #gluster-dev
05:59 ppai joined #gluster-dev
06:01 spalai1 joined #gluster-dev
06:02 ankitraj joined #gluster-dev
06:02 hgowtham joined #gluster-dev
06:15 kdhananjay joined #gluster-dev
06:31 pranithk1 joined #gluster-dev
06:47 asengupt joined #gluster-dev
06:50 ndevos aravindavk: there is a problem with the packaging of that file, it's part of -server *and* -events, it should only be in the last
06:50 ndevos I;ve left a comment on how to fi that
06:50 ndevos s/fi /fix /
06:50 aravindavk ndevos: thanks
06:55 EinstCrazy joined #gluster-dev
06:58 EinstCra_ joined #gluster-dev
06:59 ffilzwin2 joined #gluster-dev
07:01 pranithk1 xavih: hey, just had one comment on aspandey's patch. I fixed that problem as well once I spoke to him. Let us know what you think of that part of the change.
07:01 xavih pranithk1: I'm just looking at it :)
07:02 pranithk1 xavih: great. Once this is merged, aspandey will start working on heal-info
07:02 pranithk1 aspandey: please don't port this to 3.9 until heal-info work is also done...
07:02 aspandey pranithk1, ok..
07:04 xavih pranithk1: I have a doubt...
07:04 spalai joined #gluster-dev
07:05 xavih pranithk1: the xattrop is sent as a subfop of the main fop. This means that any failure in the xattrop itself will be propagated to the main fop (update of the good mask)
07:06 pranithk1 xavih: true
07:06 pranithk1 xavih: In which case marking of dirty may or may not happen, which is fine
07:06 xavih pranithk1: the xattrop itself will only succeed if at least k bricks succeed. If it fails, there could be some bricks that succeeded, so the file should be dirty anyway
07:07 pranithk1 xavih: yes
07:08 xavih pranithk1: I'm not sure if it's really necessary to update ctx->dirty once the xattrop succeeded
07:09 pranithk1 xavih: How do we remember we don't need to keep doing marking dirty again and again without it?
07:09 pranithk1 xavih: ah! you mean we can update it before?
07:10 pranithk1 xavih: The problem I thought of is that, if all the bricks give disconnect at the same time or something, it would still think ctx->dirty is set so it will try to do -1, at that time if the bricks are up, dirty will be set to -1
07:10 pranithk1 xavih: so these kinds of variety things will happen
07:11 xavih pranithk1: I see...
07:12 xavih pranithk1: ok, anyway this change doesn't hurt until we allow updating changes to be processed in parallel...
07:12 xavih pranithk1: I'll think about this when the change is made :)
07:12 xavih pranithk1: current solution seems ok to me
07:13 pranithk1 xavih: cool, then It is a +2 from me with this change. Feel free to merge it if you see that all the regressions pass in your day time.
07:13 pranithk1 xavih: Otherwise I will do it off
07:13 xavih pranithk1: one more question...
07:13 pranithk1 xavih: yes, tell me
07:13 xavih pranithk1: let me see the code before speaking... :P
07:14 pranithk1 xavih: sure :-)
07:15 xavih pranithk1: It seems tha dirty xattr is handled exactly as version xattr, right ?
07:15 rafi1 joined #gluster-dev
07:16 pranithk1 xavih: No, version xattr has to match on at least k bricks
07:16 pranithk1 xavih: not for dirty
07:16 pranithk1 xavih: as long as the xattrop happens on at least k bricks we are fine
07:17 pranithk1 xavih: the dirty count may not match
07:17 pranithk1 xavih: at least that is my understanding...
07:18 xavih pranithk1: right. I have to refresh many thing after holidays... however, it seems that cbk->dirty is not used anywhere now...
07:19 pranithk1 xavih: yeah, I see that the regression is failing in ec-background-heals.t, I will take a look...
07:20 xavih pranithk1: I'll analyze this later. For now the patch is ok
07:20 pranithk1 xavih: okay
07:22 xavih pranithk1: the first failed test is a kill_brick...
07:22 xavih pranithk1: it's weird to fail there...
07:23 pranithk1 xavih: I will check, don't worry
07:23 pranithk1 xavih: may be I didn't handle some case, I only tested one case before sending the patch because I wasn't sure everyone will accept this change
07:24 pranithk1 aravindavk: hey, events in 'init' are not coming up in the events. Is this expected?
07:25 pranithk1 aravindavk: when is gf_event enabled?
07:25 aravindavk pranithk1: init?
07:25 pranithk1 aravindavk: I mean xlator init
07:25 xavih pranithk1: sorry, I was looking at an old version of the script. It's failing waiting for heal to finish...
07:26 aravindavk pranithk1: is event added using gf_event?
07:26 pranithk1 xavih: I will take a look man, I will make sure things work fine and resubmit this.
07:26 pranithk1 aravindavk: yes
07:27 aravindavk pranithk1: do you see any issue with format? in /var/log/glusterfs/events.log
07:27 EinstCrazy joined #gluster-dev
07:27 aravindavk pranithk1: if glustereventsd is running then it should get it
07:27 pranithk1 aravindavk: let me check
07:28 pranithk1 aravindavk: I don't have that file...
07:28 pranithk1 aravindavk: Other events came, only the events in 'init' of the xlator didn't come
07:31 aravindavk pranithk1: ok, let me check. is it merged patch?
07:31 pranithk1 aravindavk: no, http://review.gluster.org/#/c/15342/3
07:32 pranithk1 aravindavk: in posix.c, we have all these events in 'init()' which were not coming
07:33 aravindavk joined #gluster-dev
07:33 pranithk1 aravindavk: hey
07:33 pranithk1 aravindavk: found the bug
07:33 pranithk1 aravindavk: nothing to do with gf_event, sorry for wasting your time :-(
07:33 aravindavk pranithk1: np. :)
07:34 pranithk1 itisravi: kdhananjay: Anyone of you reviewing http://review.gluster.org/#/c/15378/
07:35 itisravi pranithk1: I can take a look. poornima and I were discussing about the issue
07:36 pranithk1 itisravi: cool
07:40 EinstCrazy joined #gluster-dev
07:51 rafi1 joined #gluster-dev
07:58 pranithk1 aravindavk: Is it not advised to put strings for events? Something like this: "2016-09-14 13:23:37  POSIX_BRICK_VERIFICATION_FAILED 00e57ead-06cf-4741-9404-37051b912b65 msg=Not able to determine volume brick=dhcp35-190.lab.eng.bl​r.redhat.com:/home/gfs/r2_0
07:58 pranithk1 "
08:03 spalai1 joined #gluster-dev
08:03 aravindavk pranithk1: not much use if it is descriptive. It is meant to be consumed programatically. Key value pairs makes more sense
08:04 pranithk1 aravindavk: hmm...
08:10 spalai1 left #gluster-dev
08:21 aspandey joined #gluster-dev
08:22 pranithk1 aravindavk: Is there a possibility for the key value pairs to be rearranged in the event output?
08:22 pranithk1 aravindavk: I got "2016-09-14 13:49:53  POSIX_SAME_GFID      00e57ead-06cf-4741-9404-37051b912b65 gfid=a3123b8b-2661-4a89-8662-c3048a437da6 newpath=/a brick=dhcp35-190.lab.eng.bl​r.redhat.com:/home/gfs/r2_1 path=/home/gfs/r2_1/.glusterfs/a3/12/a​3123b8b-2661-4a89-8662-c3048a437da6", but this is not the order I gave them in the gf_event()
08:22 pranithk1 aravindavk: 'brick' was supposed to be last
08:35 gem joined #gluster-dev
08:35 atinm joined #gluster-dev
08:41 pranithk1 atinm: done with posix-events man...
08:42 nigelb kshlm: If you've got a minute, http://review.gluster.org/#/c/15495/
08:42 nigelb The tests were returning zero but reporting to gerrit as failure when build fails. That sounds wrong.
08:43 atinm pranithk1, cool, backport to 3.9 :)
08:43 pranithk1 atinm: yeah
08:45 devyani7 joined #gluster-dev
08:53 pranithk1 atinm: http://review.gluster.org/15497
08:54 atinm pranithk1, mainline is yet to be merged right?
08:54 pranithk1 atinm: yeah
08:58 aravindavk pranithk1: json data sorted by key, that should be fine
09:03 aravindavk pranithk1: json data for the same will look like {"ts": 1473841193, "event": "POSIX_SAME_GFID", "nodeid": "00e57ead-06cf-4741-9404-37051b912b65", "message": {"gfid": "a3123b8b-2661-4a89-8662-c3048a437da6", "newpath": "/a", "brick": "dhcp35-190.lab.eng.blr.redhat.com:/home/gfs/r2_1" ,"path":"/home/gfs/r2_1/.glusterfs/a3/12​/a3123b8b-2661-4a89-8662-c3048a437da6"}}
09:04 pranithk1 aravindavk: cool, guessed so, sent out the patch
09:04 pranithk1 aravindavk: thanks a lot for all the help
09:04 aravindavk pranithk1: welcome
09:08 kshlm nigelb, Done.
09:08 pranithk1 joined #gluster-dev
09:11 nigelb kshlm: thanks.
09:15 misc trying to find my way in the RHEL 6 channels: https://www.youtube.com/watch?v=JtEkUmYecnk
09:16 nigelb I'm wondering how meta it would be to have Jenkins update jobs on it's own.
09:17 misc mhh meta, yes
09:17 misc but I am not sure if jenkins updating his config is safe
09:18 spalai joined #gluster-dev
09:19 spalai1 joined #gluster-dev
09:20 xavih joined #gluster-dev
09:23 poornima joined #gluster-dev
09:30 atinm joined #gluster-dev
09:40 aravindavk ppai: hi
09:42 aravindavk ppai: https://github.com/gluster/glusterdocs/blob/m​aster/Administrator%20Guide/Events%20APIs.md is merged but respective page is not found in readthedocs http://gluster.readthedocs.io/en/latest​/Administrator%20Guide/Events%20APIs.md
09:42 aravindavk ppai: am I missing anything?
09:42 nbalacha joined #gluster-dev
09:42 mchangir joined #gluster-dev
09:43 rastar joined #gluster-dev
09:46 aravindavk ppai: entry in mkdocs.yml?
09:56 spalai1 nigelb: ping, 02:34:01 ./tests/basic/gfapi/gfapi-ssl-test.c:10:40: error: glusterfs/api/glfs-handles.h: No such file or directory
09:56 spalai1 nigelb: testcase  : ./tests/basic/gfapi/gfapi-ssl-test.t failed thrice
09:58 aravindavk ppai: got it, added to mkdocs.yml and it is working now https://github.com/gluster/glusterdocs/pull/149
09:59 nigelb aravindavk: ah, sorry about that. The tests accidentally got disabled.
10:00 nigelb spalai1: release-3.8
10:00 nigelb ?
10:00 spalai1 nigelb: yes
10:00 spalai1 https://build.gluster.org/job/ce​ntos6-regression/724/consoleFull
10:00 aravindavk nigelb: which tests?
10:00 nigelb aravindavk: there are travis tests running for docs.
10:00 nigelb which somehow got disabled.
10:01 aravindavk nigelb: it says all checks completed https://github.com/gluster/glusterdocs/pull/149
10:02 nigelb it does now.
10:02 nigelb YOur original pull request should have had tests.
10:02 nigelb It didn't.
10:03 aravindavk nigelb: ok
10:03 nigelb spalai1: mohit also emailed me about ssh tests failing on release-3.8
10:03 nigelb I think release-3.8 is busted.
10:03 nigelb talk to Rajesh?
10:05 ppai aravindavk: ok, looking now
10:06 aravindavk ppai: sent pull request now, Checked locally. adding to mkdocs.yml did the trick
10:06 ppai aravindavk: okay, I'll merge it
10:06 aravindavk ppai: thanks
10:06 spalai1 nigelb: Rajesh asked me to consult you
10:07 spalai1 nigelb: he pointed there might be issues with machine
10:08 aravindavk ppai++
10:08 glusterbot aravindavk: ppai's karma is now 17
10:08 nigelb spalai1: link to the failure?
10:08 nigelb or the change
10:08 spalai1 nigelb: https://build.gluster.org/job/ce​ntos6-regression/724/consoleFull
10:09 spalai1 nigelb: https://build.gluster.org/job/ce​ntos6-regression/740/consoleFull
10:09 nigelb When we last had infra issues, the problem was that the keys were not being generated.
10:10 nigelb This time, the keys are generated.
10:12 nigelb spalai1: File a bug. I'll dig. I'm about 80% sure it's a code issue and not infra.
10:12 nigelb I'll just have to get some proof for my theory :)
10:13 spalai1 nigelb: here are the logs
10:13 spalai1 ./tests/basic/gfapi/gfapi-ssl-test.c:9:32: error: glusterfs/api/glfs.h: No such file or directory 04:11:47 ./tests/basic/gfapi/gfapi-ssl-test.c:10:40: error: glusterfs/api/glfs-handles.h: No such file or directory 04:11:47 ./tests/basic/gfapi/gfapi-ssl-test.c:16: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘*’ token
10:14 spalai1 nigelb: it could not find glfs.h, gfapi-ssl-test.c
10:14 spalai1 nigelb: some issue with path
10:17 nishanth joined #gluster-dev
10:21 rraja joined #gluster-dev
10:24 nbalacha joined #gluster-dev
10:27 spalai1 nigelb: ^^
10:28 spalai1 nigelb: filed the bug here: https://bugzilla.redhat.co​m/show_bug.cgi?id=1375952
10:28 glusterbot Bug 1375952: unspecified, unspecified, ---, bugs, NEW , Regression failure for test case: ./tests/basic/gfapi/gfapi-ssl-test.t
10:30 EinstCrazy joined #gluster-dev
10:50 rastar joined #gluster-dev
10:51 poornima joined #gluster-dev
10:53 nigelb rastar: Is there a guide somewhere of how the test framework works?
11:05 nigelb poornima: do you have a minute to help me navigate the gfapi tests?
11:05 nigelb (see bug 1375952)
11:05 glusterbot Bug https://bugzilla.redhat.com:​443/show_bug.cgi?id=1375952 unspecified, unspecified, ---, bugs, NEW , Regression failure for test case: ./tests/basic/gfapi/gfapi-ssl-test.t
11:06 nigelb something is throwing a compilation error, which I suspect is a path issue
11:07 ashiq joined #gluster-dev
11:08 jiffin1 joined #gluster-dev
11:14 nigelb spalai1: code issue. I'm commenting on the bug.
11:14 nigelb post-factum++
11:14 glusterbot nigelb: post-factum's karma is now 30
11:14 nigelb thanks for writing good comments in your commits :)
11:16 post-factum hmmmm
11:17 post-factum didn't get the point but thx ;)
11:17 nigelb post-factum: so I ran into gfapi test failures on 3.8
11:17 nigelb You wrote a detailed comment when change a test from .t to .sh on why it's been changed.
11:17 post-factum not me
11:18 post-factum iirc, ndevis fixed it
11:18 post-factum i mmight be wrong
11:18 post-factum plus ndevos just to be sure :)
11:18 nigelb http://review.gluster.org/#/c/15223/
11:18 nigelb I saw your name in author. I didn't actually read the review.
11:19 post-factum yep, ndevos++
11:19 glusterbot post-factum: ndevos's karma is now 311
11:22 nigelb post-factum: so, long story short, one more .t file slipped through.
11:22 nigelb so release-3.8 is effectively broken right now.
11:24 nigelb ndevos: Another .t file slipped through into gfapi tests, so patches into 3.8 are blocked by test failure. I've filed bug 1375952 and assigned it to the right person.
11:24 glusterbot Bug https://bugzilla.redhat.com:​443/show_bug.cgi?id=1375952 unspecified, unspecified, ---, bugs, NEW , Regression failure for test case: ./tests/basic/gfapi/gfapi-ssl-test.t
11:24 post-factum nigelb: no, wait
11:24 post-factum nigelb: the fix should be merged into 3.8
11:25 nigelb post-factum: A new file got backported.
11:26 nigelb so that's gone and done the same thing again.
11:28 post-factum nigelb: ah, that is effectively *another* test
11:28 post-factum nigelb: the yes, my fix should be useful
11:28 post-factum *then
11:28 nigelb yeah.
11:28 nigelb thanks to your commit, I figured out *why* it was breaking :)
11:29 post-factum heh, in fact, at first 3.8 *passed* that test, and that is why i missed it, mut then it broke everything, and i had to make a fix
11:29 post-factum dunno why it passed at the beginning
11:30 nigelb It occasionally passes.
11:30 nigelb if you brute force it enough.
11:30 nigelb Even now.
11:30 nigelb I have no idea how or why.
11:30 post-factum at the same time, 3.7 broke immediately, and i fixed it within initial commit
11:42 ankitraj joined #gluster-dev
11:44 Saravanakmr joined #gluster-dev
11:48 ankitraj joined #gluster-dev
11:48 spalai1 nigelb++
11:48 glusterbot spalai1: nigelb's karma is now 30
11:48 gem joined #gluster-dev
12:00 jdarcy joined #gluster-dev
12:01 ankitraj #info #startmeeting Gluster Community Meeting
12:06 xavih joined #gluster-dev
12:06 rastar joined #gluster-dev
12:11 kotreshhr1 joined #gluster-dev
12:13 nishanth joined #gluster-dev
12:18 mchangir who is the 3.8.4 release manager please?
12:20 mchangir aravindavk, ^^
12:21 kshlm mchangir, ndevos
12:22 mchangir kshlm, thanks
12:40 mchangir joined #gluster-dev
12:43 kotreshhr joined #gluster-dev
12:43 shyam joined #gluster-dev
12:43 ashiq joined #gluster-dev
12:46 jiffin joined #gluster-dev
12:58 kshlm samikshan++
12:58 glusterbot kshlm: samikshan's karma is now 4
13:07 nbalacha joined #gluster-dev
13:08 mchangir joined #gluster-dev
13:09 ira joined #gluster-dev
13:10 atinm joined #gluster-dev
13:15 jiffin joined #gluster-dev
13:16 spalai1 left #gluster-dev
13:21 shyam joined #gluster-dev
13:31 kshlm joined #gluster-dev
13:52 nbalacha joined #gluster-dev
13:56 nigelb nbalacha: what ended up being the conclusion to the freebsd thing?
13:56 nigelb Do we end up not supporting UFS/FFS completely?
13:57 nbalacha nigelb, I would recommend that we do not
13:57 nbalacha unless there is someone who knows FreeBSD
13:57 nbalacha and can help us with that
13:58 nigelb If it affects NetBSD as well, might be worth CCing Emmanual into the email.
13:58 nigelb At least he somewhat actively helps out.
14:06 kotreshhr left #gluster-dev
14:14 shyam joined #gluster-dev
14:34 nbalacha I sent it to gluster-devel
14:34 nbalacha so I guess he might have seen it?
14:40 nigelb I kinda want to give that thread a bump. Whether we support *BSD or not is relevant to my interests :)
14:43 nigelb ndevos: when you've got time, I could use some help understanding what's going on in this failure -  https://build.gluster.org/job/cage-test/1/console
14:50 hagarth joined #gluster-dev
14:51 Muthu_ joined #gluster-dev
15:14 shyam joined #gluster-dev
15:24 wushudoin joined #gluster-dev
16:23 gem joined #gluster-dev
16:36 hagarth joined #gluster-dev
16:36 shyam joined #gluster-dev
16:37 ndevos nigelb: maybe rpcbind is not enabled+running?
16:37 ndevos nigelb: also make sure that nfs-utils is installed
16:39 nigelb ah, nfs-utils might be the missing piece
16:39 nigelb thanks
16:41 nigelb nope, rpcbind was not running or enabled.
16:42 misc so we need to have it started ?
16:43 misc (if you tell me the failure, I will add that to ansible)
16:43 misc should be in roles/jenkins_builder/ , and I can take a look after the docker meetup
16:45 ndevos I *think* it should get started by the tests too, but maybe that is not done...
17:07 shyam joined #gluster-dev
17:08 glusterbot joined #gluster-dev
17:08 skoduri joined #gluster-dev
17:13 jiffin joined #gluster-dev
17:27 rafi joined #gluster-dev
17:44 nigelb misc: I'll give you a pull request
17:44 nigelb ... after I confirm it works.
17:58 shyam joined #gluster-dev
18:12 nigelb ndevos: more errors to debug - mount.nfs: access denied by server while mounting builder1.rht.gluster.org:/patchy
18:12 nigelb See https://build.gluster.org/job/cage-test/2/console
18:39 rafi joined #gluster-dev
18:42 jiffin joined #gluster-dev
20:20 glusterbot joined #gluster-dev
20:22 hagarth joined #gluster-dev
20:23 shyam joined #gluster-dev
20:28 hagarth joined #gluster-dev
20:42 pranithk1 joined #gluster-dev
21:17 hagarth joined #gluster-dev
21:54 hagarth1 joined #gluster-dev

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary