Please see below…
-----Original Message----- From: Michael C. Jaeger [mailto:m...@mcj.de] Sent: Tuesday, February 5, 2019 12:51 PM To: Kline, Charles (US) <charles.kl...@lmco.com> Cc: Jaeger, Michael C. <michael.c.jae...@siemens.com>; Stangel, Dan <dan.stan...@hpe.com>; fossol...@fossology.org Subject: Re: EXTERNAL: Re: [FOSSology] Quick Question Hi, so did not like the idea of segmentation. I am also not sure what was going on. A geneal place to look at is the apache log sudo tail -n 1000 /var/log/apache2/error.log >>>>> /var/log/apache2, apache or tomcat do not exist. or the fossology log sudo tail -n 1000 /var/log/fossology/fossology.log >>>>> we have /app/fossology/log/fossology where fossology is a file. While I do see ERROR messages (e.g. “xxx_pffile already exists”, FATAL messages like for the database for connection issues…it has been a while since there have been any issues…for the most part this file is showing Fossology scheduler started or closed messages. maybe something is written there which gives to you some hint about what is going on your machine. Another thing is that you mention only two php.ini settings changed, but also a third one requires changes: memory_limit >>> I checked this out in the php.ini file…it is currently set to 128M. please see more details here: https://github.com/fossology/fossology/blob/master/install/scripts/php-conf-fix.sh >>> I checked this out…I recall seeing this in my wanderings through the fossology sites, per this site I see the following that I assume possibly default OOTB settings maybe??? if [ -e $phpIni ] then echo 'Copies php.ini to current directory and creates a backup file' echo 'Modifies it and then displays variance for confirmation.' cp $phpIni php.ini.orig cp php.ini.orig php.ini echo 'If happy with changes you should save original and move new one back.' echo 'Setting max execution time to 300 seconds (5 mins)' sed -i 's/^\(max_execution_time\s*=\s*\).*$/\1300/' php.ini echo 'Setting memory limit to 702M' sed -i 's/^\(memory_limit\s*=\s*\).*$/\1702M/' php.ini echo 'Setting post max size to 701M' sed -i 's/^\(post_max_size\s*=\s*\).*$/\1701M/' php.ini echo 'Setting max upload filesize to 700M' sed -i 's/^\(upload_max_filesize\s*=\s*\).*$/\1700M/' php.ini echo "Setting timezone to $TIMEZONE" sed -i "s%.*date.timezone =.*%date.timezone = $TIMEZONE%" php.ini echo 'php.ini adjusted!' echo echo 'Display the changes made' diff php.ini.orig php.ini echo $1 If I look at our /etc/php.ini I see the following: Max_execution_time: 300 >>> do you see a need for increasing this value. I understand one needs to be careful with this setting in particular on production servers so you don’t have unexpectedly long running scripts that could potentially impact other jobs/users. Memory_limit: 702M >>>>>>>>>>>>> Do you have a suggestion as to what this should be increased to???? Should it be the same as post_max_size??? Post_max_size: 4096M (we had manually updated the php.ini file to increase this from 2048M) Upload_max_filesize: 4096M (we had manually updated the php.ini file to increase this from 2048M) Kind regards, Michael > On 4. Feb 2019, at 21:22, Kline, Charles > <charles.kl...@lmco.com<mailto:charles.kl...@lmco.com>> wrote: > > Hi guys...this is killing me! > Another question on uploads...currently php.ini has been updated setting > upload_max_filesize and post_max_size to 4096M and on the Upload a New File > screen I see " FOSSology server has imposed a maximum upload file size of > 4096M bytes."...which is interesting because last week after making the > property updates I still saw it indicating "2048M bytes". I thought maybe a > server reboot had been done while I was out...but using 'uptime' indicates > the server has been up for 22 days. > > Anyways...the user tried uploading a 2.8G file using Chrome. When initiated > she saw the %complete in the lower left hand corner and the spinning thing at > the upper left hand corner which they always see. The upload reached 100% > and the spinning stopped but no job id was generated with the link you could > click on despite waiting a while. We are at a loss as to what is going > on...mostly from our general ignorance. Not sure if at this point it is > determining there isn't sufficient RAM to process the uploaded file (I am > currently trying to get the RAM increased on this box from 8G to 24G) or > what. There would appear to be plenty of actual storage available. Any > thoughts on what is going on at this point and why we are not processing the > file that appeared to upload successfully? > > Charlie > > -----Original Message----- > From: Kline, Charles (US) > Sent: Thursday, January 31, 2019 8:26 AM > To: 'Michael C. Jaeger' <m...@mcj.de<mailto:m...@mcj.de>> > Cc: 'Jaeger, Michael C.' > <michael.c.jae...@siemens.com<mailto:michael.c.jae...@siemens.com>>; 'Stangel, > Dan' <dan.stan...@hpe.com<mailto:dan.stan...@hpe.com>>; > 'fossol...@fossology.org' > <fossol...@fossology.org<mailto:fossol...@fossology.org>> > Subject: RE: EXTERNAL: Re: [FOSSology] Quick Question > > Tried an upload using Chrome...I like that! Thanks...I have passed this on so > someone can test a large file using Chrome. > > Hey, any thoughts on this...this was posed by one of our administrators..." > The location where the file is uploaded to, does that have enough space to > read the file. I don’t know if fossoloy uploads to a local area first and > then puts the file in the define location so I can’t say where this location > would be." > > While trying to test last night with >2G files I was running 'top -u > fossy'...as usual the only running was fo_scheduler...but this also show MEM > and SWAP....the Available memory showed 8G...the used showed 7.xG with ~500M > freeI never saw the free space fall much below this and I often see these > numbers. Later when testing a very very small file I did see the Used in the > 5G range. > > Charlie > > -----Original Message----- > From: Kline, Charles (US) > Sent: Thursday, January 31, 2019 8:14 AM > To: 'Michael C. Jaeger' <m...@mcj.de<mailto:m...@mcj.de>> > Cc: Jaeger, Michael C. > <michael.c.jae...@siemens.com<mailto:michael.c.jae...@siemens.com>>; Stangel, > Dan > <dan.stan...@hpe.com<mailto:dan.stan...@hpe.com>>; > fossol...@fossology.org<mailto:fossol...@fossology.org> > Subject: RE: EXTERNAL: Re: [FOSSology] Quick Question > > Hi Michael, > Yeah, big thing is we are trying to ascertain whether or not the file is > actually uploaded...certainly the test file I can see uploaded as a job > number was assigned and the ununpack started. I'll check out Chrome. > > charlie > > -----Original Message----- > From: Michael C. Jaeger [mailto:m...@mcj.de] > Sent: Thursday, January 31, 2019 7:10 AM > To: Kline, Charles (US) > <charles.kl...@lmco.com<mailto:charles.kl...@lmco.com>> > Cc: Jaeger, Michael C. > <michael.c.jae...@siemens.com<mailto:michael.c.jae...@siemens.com>>; Stangel, > Dan > <dan.stan...@hpe.com<mailto:dan.stan...@hpe.com>>; > fossol...@fossology.org<mailto:fossol...@fossology.org> > Subject: Re: EXTERNAL: Re: [FOSSology] Quick Question > > Hello, > > looks good. it appears also to me that a part of your waiting time appears to > be the upload / transfer time. Not all browsers show the upload progress, > Chrome does in the lower left corner. On some OSes you can see the network > traffic. I think this is maybe also another effect when dealing with VLP- > "very large packages" Maybe FOSSology could improve on this by having an > upload status dialogue to inform the user more precisely. > > I think the last two screenshots show the progress info after the file has > been uploaded to the FOSSology server. > > Kind regards, Michael > >> On 31. Jan 2019, at 09:53, Kline, Charles >> <charles.kl...@lmco.com<mailto:charles.kl...@lmco.com>> wrote: >> >> Hmmm, not sure I’m following you Michael….here’s some screenshots >> from a test I did… <image003.png><image011.jpg> CLICK UPLOAD… >> <image012.png><image004.png> Then I will see this…but this is after >> the uploaded has been completed and the ununpack has started as >> shown below… <image014.png><image008.png> >> <image015.png><image010.png> I guess I was hoping there was a file or log on >> the fossology server that one could tail to see the progress or at least >> repeatedly do a listing to see the file growth. >> >> Charlie >> >> From: Jaeger, Michael C. [mailto:michael.c.jae...@siemens.com] >> Sent: Thursday, January 31, 2019 1:50 AM >> To: Kline, Charles (US) >> <charles.kl...@lmco.com<mailto:charles.kl...@lmco.com>>; Michael C. Jaeger >> <m...@mcj.de<mailto:m...@mcj.de>> >> Cc: Stangel, Dan <dan.stan...@hpe.com<mailto:dan.stan...@hpe.com>>; >> fossol...@fossology.org<mailto:fossol...@fossology.org> >> Subject: RE: EXTERNAL: Re: [FOSSology] Quick Question >> >> Hi, >> >> If you upload a file, there will be a link presented in the “message bar” >> right below the menus with the number (id) of the upload. Clicking this >> number will show you a table with the progress. >> >> If you have missed that you could click on “Jobs” in the top menu bar, as >> for the options choose “my recent jobs”, it should show what is currently >> ongoing. >> >> Kind regards, Michael >> >> From: Kline, Charles [mailto:charles.kl...@lmco.com] >> Sent: Donnerstag, 31. Januar 2019 00:53 >> To: Jaeger, Michael C. (CT RDA SSI); Michael C. Jaeger >> Cc: Stangel, Dan; fossol...@fossology.org<mailto:fossol...@fossology.org> >> Subject: RE: EXTERNAL: Re: [FOSSology] Quick Question >> >> Hi, >> Is there a way to monitor (whether it be via the fossology GUI or on the >> fossology server) the progress of the upload of a file? Is there a log or >> something that can be monitored? In a previous test with a 2G file it took >> ~10 min to upload. I don’t know if this is a linear thing where a 4G file >> will take 20 minutes, but it would certainly be nice if one could tell how >> far along the upload was so you don’t abort the test prematurely because you >> felt you had given it enough time. >> >> Thanks in advance!!!! >> >> Charlie >> >> From: Jaeger, Michael C. [mailto:michael.c.jae...@siemens.com] >> Sent: Tuesday, January 22, 2019 7:36 AM >> To: Kline, Charles (US) >> <charles.kl...@lmco.com<mailto:charles.kl...@lmco.com>>; Michael C. Jaeger >> <m...@mcj.de<mailto:m...@mcj.de>> >> Cc: Stangel, Dan <dan.stan...@hpe.com<mailto:dan.stan...@hpe.com>>; >> fossol...@fossology.org<mailto:fossol...@fossology.org> >> Subject: EXTERNAL: RE: EXTERNAL: Re: [FOSSology] Just curious...question >> about your installation... >> >> Hello, >> >> uploading 15GB in a single upload is not advisable in any case, since it >> will lead to speed issues when browsing the source in an aggregated view. >> >> 15GB of compressed source code will be like … 100GB of uncompressed files? >> that will be that will be maybe like >1000 packages into a single upload? >> maybe magnitude of 1 mio files in one upload? -> that will lead to endlessly >> running queries in the aggregated view unless you are really good at tuning >> your postgresql server. >> >> I would highly recommend to consider splitting your 15GB upload into 30 >> packages or so. >> >> Kind regards, Michael >> >> >> >> From: fossology@lists.fossology.org<mailto:fossology@lists.fossology.org> >> [mailto:fossology@lists.fossology.org] On Behalf Of Kline, Charles >> Sent: Montag, 21. Januar 2019 20:10 >> To: Michael C. Jaeger >> Cc: Stangel, Dan; fossol...@fossology.org<mailto:fossol...@fossology.org> >> Subject: Re: EXTERNAL: Re: [FOSSology] Just curious...question about your >> installation... >> >> Hi Michael, >> Yeah, this could certainly be an issue...they would like to be able >> to process a 15G file and I don’t think that is going to happen…if we >> could get 4G that would be nice, but based on the info below that may >> not even be possible. I’ll be testing tonight with the following >> php.ini properties updated to 4096M but I’m not holding out much >> hope… >> >> Maximum allowed size for uploaded files. >> upload_max_filesize = 2048M >> Maximum size of POST data that PHP will accept. >> post_max_size = 2048M >> >> Excerpt from a 'lscpu': >> Architecture: x86_64 >> CPU op-mode(s): 32-bit, 64-bit >> CPU(s): 4 >> >> If I do a ‘cat /proc/meminfo’: >> MemTotal: 8056932 kB >> MemFree: 430432 kB >> >> And not much running right now…results of ‘top –u fossy’, seems like >> we might be up against the limit now… top - 13:29:18 up 8 days, 20:01, 1 >> user, load average: 1.00, 1.38, 1.81 >> Tasks: 210 total, 2 running, 208 sleeping, 0 stopped, 0 zombie >> Cpu(s): 3.3%us, 0.8%sy, 0.0%ni, 94.8%id, 1.1%wa, 0.0%hi, 0.1%si, >> 0.0%st >> Mem: 8056932k total, 7627692k used, 429240k free, 17668k buffers >> Swap: 4193276k total, 157492k used, 4035784k free, 5471620k cached >> >> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND >> 17259 fossy 20 0 99.1m 30m 2820 R 92.3 0.4 10:01.71 nomos >> 7075 fossy 20 0 836m 2960 1668 S 0.0 0.0 1:48.17 fo_scheduler >> >> Charlie >> >> -----Original Message----- >> From: Michael C. Jaeger [mailto:m...@mcj.de] >> Sent: Friday, January 18, 2019 5:29 AM >> To: Kline, Charles (US) >> <charles.kl...@lmco.com<mailto:charles.kl...@lmco.com>> >> Cc: Stangel, Dan <dan.stan...@hpe.com<mailto:dan.stan...@hpe.com>>; >> fossol...@fossology.org<mailto:fossol...@fossology.org> >> Subject: EXTERNAL: Re: [FOSSology] Just curious...question about your >> installation... >> >> Hello, >> >> I think as for the PHP is it the question how much RAM you have (and >> ... do you run a 32-bit OS?) >> >> restarting apache depends on your linux distro, in some cases >> >> sudo service apache2 restart >> >> or a tool from the apache web server software could also help (which >> could be installed on your system, I do not know) >> >> sudo apachectl restart >> >> in other cases google will help. Reading into apachectl can help you >> to do a more careful handling, such as >> >> apachectl configtest >> apachectl graceful >> >> ... >> >> Kind regards, Michael >> >>> On 16. Jan 2019, at 20:07, Kline, Charles >>> <charles.kl...@lmco.com<mailto:charles.kl...@lmco.com>> wrote: >>> >>> Just curious...what you have in your pkp.ini file for these properties...or >>> I guess I should ask can you process files greater than 2G? >>> ; Maximum allowed size for uploaded files. >>> upload_max_filesize = 2048M >>> >>> ; Maximum size of POST data that PHP will accept. >>> post_max_size = 2048M >>> >>> Oh, do you know how to restart Apache??? Trying to figure that out. >>> >>> Charlie >>> >>> -----Original Message----- >>> From: Kline, Charles (US) >>> Sent: Wednesday, January 9, 2019 2:31 PM >>> To: 'Stangel, Dan' <dan.stan...@hpe.com<mailto:dan.stan...@hpe.com>>; >>> fossol...@fossology.org<mailto:fossol...@fossology.org> >>> Subject: RE: Question on version 2.5.0 >>> >>> Thanks so much Dan!!!! >>> >>> -----Original Message----- >>> From: Stangel, Dan [mailto:dan.stan...@hpe.com] >>> Sent: Wednesday, January 9, 2019 1:52 PM >>> To: Kline, Charles (US) >>> <charles.kl...@lmco.com<mailto:charles.kl...@lmco.com>>; >>> fossol...@fossology.org<mailto:fossol...@fossology.org> >>> Subject: EXTERNAL: RE: Question on version 2.5.0 >>> >>> Hi Charlie, >>> >>> Ah, the joys of a large enterprise IT organization. You ask a lot of good >>> questions, and I would probably defer to other experts on this list for >>> most of them, but I definitely can say that the delete agent has to run >>> exclusively, by design, and that it can be rather slow. So your users' >>> experiences are not surprising, and I'm not sure there's a good solution. >>> Upgrading to a newer version may resolve some of these issues, but not sure >>> what the team has done around delagent in the past few releases. >>> >>> As for maintenance tasks, most of these should be pretty reliable, but to >>> your point, I'm not sure what impact they have on running jobs. You would >>> definitely want to run the vacuum analyze job somewhat routinely, for the >>> health and well-being of Postgres. I think in our configuration it is set >>> up as a cron job. >>> >>> Dan >>> >>> >>> From: Kline, Charles [mailto:charles.kl...@lmco.com] >>> Sent: Wednesday, January 09, 2019 11:06 AM >>> To: Stangel, Dan <dan.stan...@hpe.com<mailto:dan.stan...@hpe.com>>; >>> fossol...@fossology.org<mailto:fossol...@fossology.org> >>> Subject: RE: Question on version 2.5.0 >>> >>> Dan, >>> At your convenience if you had any thoughts on the below I would appreciate >>> it....having said that I'm aware that I'm kinda taking advantage of you at >>> this point so please feel free to decline. I'm sure you have plenty on your >>> plate. >>> >>> Back in Nov (I was out for surgery) the users reported the following: >>> To the two server admins who kinda support the fossology server from one of >>> the main fossology users: (apparently user was trying to do some >>> deletes.assumption is that 'regular' fossology activity was taking place) >>> Can one of you folks take a look at the Fossology server? >>> It's just kind of sitting there doing nothing. >>> >>> Admin responded that postgresql and fossology had been restarted. >>> >>> Subsequent feedback from user. >>> I tried deletes again, but it's not processing them. It's not even starting >>> the job. >>> And subsequently.from another super user. >>> >>> I had a big job running that didn't finish until 11/14/2018 @ 01:13 AM. >>> >>> As near as I can tell the delete jobs will wait for current jobs to >>> complete their current step, then pause the job(s), then the delete job(s) >>> executes, then the other jobs continue on with their next step. >>> >>> Your deletion jobs have completed last night around 9:54 pm just around the >>> time my job completed the ununpack step. >>> >>> >>> >>> This kicked off talk (from a server admin who I have never worked with) of >>> updating fossology to the latest version and/or moving it to another >>> server. This was met with some push back.they finally decided to wait until >>> I returned to work.which wasn't until 12/14.I talked to the one super user >>> trying to get a sense of where things stood.she insisted that there were >>> still issues with trying to do deletes during the week and that they were >>> doing them on the weekends. >>> >>> Meeting was held on 1/7 to discuss. Two main issues. >>> 1. The large upload file size, which is being addressed. >>> 2. The issue with doing the deletes and things freezing up. >>> >>> In regards to issue 2, I definitely didn't get a sense that the super user >>> was particularly interested in upgrading fossology and in regards to >>> getting a new server we now have a mandate that anything new has to be in >>> the cloud.and at this point that is not an easy process. >>> >>> Based on some of my research I ran into the Exclusive and Nokill >>> settings.as seen in >>> https://github.com/fossology/fossology/wiki/Job-Schedulerand if I look at >>> /app/fossology/fossology_2.5/share/fossology I see the following.not sure >>> if any of this is relevant to the above. >>> Ununpack >>> ; Directives: >>> ; EXCLUSIVE: the agent cannot run concurrently with any other agent. >>> special[] = NOKILL >>> >>> buckets >>> ; Directives: >>> ; EXCLUSIVE: the agent cannot run concurrently with any other agent. >>> special[]= >>> >>> copyright >>> ; Directives: >>> ; EXCLUSIVE: the agent cannot run concurrently with any other agent. >>> special[] = >>> >>> nomos >>> ; Directives: >>> ; EXCLUSIVE: the agent cannot run concurrently with any other agent. >>> special[] = >>> >>> pkgagent >>> ; Directives: >>> ; EXCLUSIVE: the agent cannot run concurrently with any other agent. >>> special[] = >>> >>> delagent.conf >>> ; Directives: >>> ; EXCLUSIVE: the agent cannot run concurrently with any other agent. >>> special[] = EXCLUSIVE >>> special[] = NOKILL >>> >>> >>> Also, wondering what your experience is with running any of the maintenance >>> tasks.any comments would be appreciated. >>> >>> I am a little hesitant running these since I am not 100% sure what they do >>> or impact to the application when they are running.the ones I have >>> considered running are "Remove uploads with no pfiles", "Remove orphaned >>> temp tables" and "Vacuum Analyze the database". >>> >>> I am retiring later this year and hope to scrape this off my shoe sooner >>> than later. >>> >>> Thanks in advance for your time and consideration. >>> >>> Charlie >>> >>> -----Original Message----- >>> From: Kline, Charles (US) >>> Sent: Wednesday, January 9, 2019 11:44 AM >>> To: 'Stangel, Dan' <dan.stan...@hpe.com<mailto:dan.stan...@hpe.com>>; >>> fossol...@fossology.org<mailto:fossol...@fossology.org> >>> Subject: RE: Question on version 2.5.0 >>> >>> Dan, >>> Thanks so much for your response...to be honest I didn't think I would get >>> one. >>> >>> Upon further diffing I tracked down /etc/php.ini which looks to what needs >>> to be updated. I was being told that only post_max_size needed to be >>> updated but totally agree that upload_max_filesize also should be updated. >>> Trick now would be to get sudo to enable me to do this or find an admin who >>> can copy an updated version I made into /etc. >>> >>> In regards to restarting Apache...again dumb question, but does >>> apache get restarted when the scheduler is restarted or much it be >>> restarted separately? Frankly I have never done anything with >>> apache. (and the big unfortunately is when I was given fossology to >>> support I was told not to spend any time on it learning anything but >>> just to respond to issues as they arise...so every time there is an >>> issue it's an >>> adventure!) >>> >>> My plan would be to initially increase the size of these 2 properties to >>> 4096M, restart the scheduler(fossology), and test. If OK, update the >>> properties to a size to accommodate the large file they are trying to >>> process and repeat the above. I don't think I'll leave it at that size, >>> 16384M, but would probably put it back to 4096M. >>> >>> Thanks!!!!! >>> Charlie >>> >>> -----Original Message----- >>> From: Stangel, Dan [mailto:dan.stan...@hpe.com] >>> Sent: Wednesday, January 9, 2019 11:33 AM >>> To: Kline, Charles (US) >>> <charles.kl...@lmco.com<mailto:charles.kl...@lmco.com>>; >>> fossol...@fossology.org<mailto:fossol...@fossology.org> >>> Subject: EXTERNAL: RE: Question on version 2.5.0 >>> >>> Hi Charlie, >>> >>> On 2018-01-08, Kline, Charles wrote: >>>> Yeah, I know...why are we still using 2.5.0...long story. >>> >>> Don't feel too bad - we're still on 2.6.2 for our production >>> instance (again, long story.. Suffice it to say these releases have >>> a long >>> lifespan) >>> >>>> Question....users want to process/upload files greater than 2GB which >>>> my users are saying is the largest then can process. Trying to >>>> figure out the correct place to update this value. >>>> >>>> In >>>> /app/fossology/fossology_2.5/share/fossology/lib/php/common-scheduler. >>>> php >>>> there is the following... >>>> common-scheduler.php: \param $MaxSize - optional max read size, >>>> default is 2048 common-scheduler.php:function >>>> fo_scheduler_read($SchedObj, $MaxSize=2048) I assume this is in MB >>>> >>>> But in the fossology online info I am seeing this...assuming these >>>> are values for the latest version of fossology...in the link they >>>> talk about updating php.ini for >>>> apache...(http://archive15.fossology.org/projects/fossology/wiki/Sy >>>> sC >>>> o >>>> nfig/annotate/9) >>>> (...) >>>> So trying to figure out where I actually need to make an update to >>>> change the size of the file that can be uploaded/processed by the users. >>> >>> You should only need to update the upload_max_filesize and >>> post_max_size parameters in your php.ini file. In our case for >>> example, we have >>> >>> ; Maximum allowed size for uploaded files. >>> ; http://php.net/upload-max-filesize >>> upload_max_filesize = 4096M >>> >>> ; Maximum size of POST data that PHP will accept. >>> ; http://php.net/post-max-size >>> post_max_size = 4096M >>> >>> Once you have updated php.ini, just be sure to restart Apache to pick up >>> the changes. Then try to do a > 2G upload from file (ideally from a client >>> that isn't too far away, network-wise) to verify that it works. >>> >>> Dan >>> >>> >>> >>> >> >> > > > > -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#3227): https://lists.fossology.org/g/fossology/message/3227 Mute This Topic: https://lists.fossology.org/mt/29603806/21656 Group Owner: fossology+ow...@lists.fossology.org Unsubscribe: https://lists.fossology.org/g/fossology/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
