Speech:Spring 2016 Thomas Rubino Log


 * Home
 * Semesters
 * Spring 2016
 * Proposal
 * Report
 * Information - General Project Information
 * Experiments - List of speech experiments

Week Ending February 9, 2016
On 2/3/16 singed into Caesar for the first time via UNH local network. Poked around to find directories and learn structure. SSH'd to Obelix 192.168.10.3 to check MNT's and things out. Found that Obelix was assigned the wrong hostname, and did not have the /mnt/main directory mounted. Informed Systems team that documentation on website does not match current config. I was unable to access any other server as they appeared either off or not configured correctly.
 * Task:

2/6/16 Still can't seem to get into cisunix.unh.edu. access denied. Emailed team to see if they are getting in, and or about a VPN connection I heard some students mentioning.
 * Task:

2/8/16 Did some more reading on the hardware setup. Still unclear on linking of installed packages. Also doing some reading on RedHat. I've mostly worked with Debian/Ubuntu servers, so yum is new to me.
 * Task:

2/9/16 Read some more logs. Can't seem to find a red hat server to work on for playing around. Using something freely available would make learning much easier.
 * Task:

2/6/16 So far just trying to get setup, and acquainted with the systems. So figured out the VPN finally. Wonder why I had never heard of this before... Was able to get into majestix. This server seems to have the /mnt/main setup.
 * Results:

2/8/16 I've got to say seems like a horrible idea that this system is out on the internet and not at least protected by SSL. Considering we log in with our UNH credentials, and they are probably sent in clear text this should be corrected. Perhaps talk to systems team about this.
 * Results:

A bit more reading on the Wiki today perhaps since I cannot get in from home yet.Got in! Still confused about the installation of things like emacs and the /mnt/main stuff. Is everything installed on Caesar and the other servers just have it?
 * Plan:
 * Concerns:

No SSL on wiki.

Week Ending February 16, 2016
2/11/16 Worked on proposal a bit tonight. Just getting the basics in there to expand upon later. Still waiting on systems team to stand up new servers so we can jump into testing and implementing some of the enhancements we talked about. Also logged into Caesar and made some changes to my $PATH to include /usr/sbin.
 * Task:

Figured out how the mounting is done on/mnt/main/local to give the other servers access to commands. A start to proposal.
 * Results:

2/14/16 A bit more to proposal. It's been a few years since I worked on linux servers, so refreshed on rsync. At one point I managed 10 or so Debian based servers and had them all backing up via rsync with and symbolic links. More about that, pulled out an old server I had in the basement hope to fire that up at work tomorrow and poke around to review my scripts and things.
 * Task:

2/14/16 Logged into Caesar to check on free space. [tsj22@caesar ~]$ df -h
 * Task:

Filesystem                    Size  Used Avail Use% Mounted on /dev/mapper/vg_caesar-lv_root   50G  3.9G   43G   9% / tmpfs                          12G   80K   12G   1% /dev/shm /dev/sda1                     485M   40M  420M   9% /boot /dev/mapper/vg_caesar-lv_home  50G  180M   47G   1% /home /dev/mapper/vg_caesar-lv_main 3.5T  624G  2.7T  19% /mnt/main

Based on this it looks like a the main logical volume might be the best spot for backups of system.

--- Logical volume --- LV Path               /dev/vg_caesar/lv_main LV Name               lv_main VG Name               vg_caesar LV UUID               5TB34g-oCjL-X96w-4cqZ-1aSi-8U45-WyzuqH LV Write Access       read/write LV Creation host, time brutus, 2015-03-03 12:45:14 -0500 LV Status             available # open                1 LV Size               3.52 TiB Current LE            923904 Segments              1 Allocation            inherit Read ahead sectors    auto - currently set to    256 Block device          253:3

I was able to get to majestic today. 192.168.1.6 and brutus.

Slightly annoying that $PATH has been altered. I assume to prevent folks from trying to run things they are not familiar with. Funny this is not the case on Brutus.

Get back into the groove of rsync, touch base with team members over next few days to figure out where we are on Experiment, and Sphinx changes.
 * Plan:

Move proposal doc back to wiki... File deleted from UNH OneDrive, Recycle Bin, and second stage Recycle bin, yeah is OneDrive great!


 * Concerns: Still not super clear on experiments. Hope to have time to read a bit Tuesday night, and chat with team.

Nigel attempting a train tonight.

Week Ending February 23, 2016
2-17-16 Update https://foss.unh.edu/projects/index.php/Speech:Function page. To include current SOX information. Repair sox
 * Task:

Fixed all missing SOX dependencies.
 * Results:

--- Using rsync --- ssh root@brutus
 * 1) Step 1.

rsync -avn --ignore-existing /usr/lib/libpulse* root@caesar:/usr/lib/
 * 1) Step 2.

# --ignore-existing makes sure not to make changes to anything already on caesar. NO FILES ON CAESAR WERE OVERWRITTEN
 * 1) This will find all files starting with libpulse and compare to those that exist on caesar.
 * 2) Using -n makes rsync run in dry mode (don't do anything)
 * 1) -v makes the output verbose
 * 2) Once all changes look ok, run without -n
 * 3) All matching files that are missing on target are copied. Including permissions.

2-18-16 Prof said sox broken again. Logged back in and fixed dependencies again. Not sure what happened, but I will take a peak at logs perhaps.
 * Task:

2-18-16 More work tonight. Setup some basic backup stuff for a few important folders /etc /mnt/main/local /mnt/main/bin /usr/lib
 * Task:

These were merely a test as it's been awhile since I played with this stuff.

Working with professor to cleanup missing files and perhaps move to a new location on /mnt/main

Next step after that will be putting the finishing touches on rsync backups and scheduling with cron so we have snapshots of every directory.
 * Plan:

These should probably live on a separate partition.

I will document entire process once it is finalized.

2-21-16 Logged in again and made some more changes to rsync setup. Added /mnt/main/scripts to new backup system
 * Task:

System still no automated, but started a README.txt at the top of /mnt/main/rsyncBackups with a detailed explanation of the systems current state and workings.

Did some more work on proposal, hope to finish up with team mates tonight. Sent email to everyone about trying to unify the formatting and contents.

Should have rsync backups on separate partition, or even machine...
 * Concerns:

Week Ending March 1, 2016

 * Task:

--- Document sox undo ---

All files compared between Brutus and Caesar.


 * /usr/lib/libvorbisenc.so.2
 * /usr/lib/libvorbisenc.so.2.0.6
 * /usr/lib/libvorbisfile.so.3
 * /usr/lib/libvorbisfile.so.3.3.2
 * /usr/lib/libvorbis.so.0
 * /usr/lib/libvorbis.so.0.4.3
 * /usr/lib/libogg.so.0
 * /usr/lib/libogg.so.0.6.0
 * /usr/lib/libwavpack.so.1
 * /usr/lib/libwavpack.so.1.0.3
 * /usr/lib/libsndfile.so.1
 * /usr/lib/libsndfile.so.1.0.20
 * /usr/lib/libpulse-browse.so.0
 * /usr/lib/libpulse-browse.so.0.1.1
 * /usr/lib/libpulsecommon-0.9.21.so
 * /usr/lib/libpulsecore-0.9.21.so
 * /usr/lib/libpulsedsp.so
 * /usr/lib/libpulse-mainloop-glib.so.0
 * /usr/lib/libpulse-mainloop-glib.so.0.0.4
 * /usr/lib/libpulse-simple.so.0
 * /usr/lib/libpulse-simple.so.0.0.3
 * /usr/lib/libpulse.so.0
 * /usr/lib/libpulse.so.0.12.2
 * /usr/lib/libX11.so.6
 * /usr/lib/libX11.so.6.3.0
 * /usr/lib/libX11-xcb.so.1
 * /usr/lib/libX11-xcb.so.1.0.0
 * /usr/lib/libdbus-glib-1.so.2
 * /usr/lib/libdbus-glib-1.so.2.1.0
 * /lib/libdbus-1.so.3
 * /lib/libdbus-1.so.3.4.0
 * /usr/lib/libSM.so.6
 * /usr/lib/libSM.so.6.0.1
 * /usr/lib/libICE.so.6
 * /usr/lib/libICE.so.6.3.0
 * /usr/lib/libXt.so.6
 * /usr/lib/libXt.so.6.0.0
 * /usr/lib/libXtst.so.6
 * /usr/lib/libXtst.so.6.1.0
 * /lib/libwrap.so.0
 * /lib/libwrap.so.0.7.6
 * /usr/lib/libgdbm_compat.so.3
 * /usr/lib/libgdbm_compat.so.3.0.0
 * /usr/lib/libgdbm.so.3
 * /usr/lib/libgdbm.so.3.0.0
 * /lib/libuuid.so.1
 * /lib/libuuid.so.1.3.0
 * /usr/lib/libX11.so.6
 * /usr/lib/libX11.so.6.3.0
 * /usr/lib/libX11-xcb.so.1
 * /usr/lib/libX11-xcb.so.1.0.0
 * /usr/lib/libXau.so.6
 * /usr/lib/libXau.so.6.0.0
 * /usr/lib/libXaw6.so.6
 * /usr/lib/libXaw6.so.6.0.1
 * /usr/lib/libXaw7.so.7
 * /usr/lib/libXaw7.so.7.0.0
 * /usr/lib/libXaw8.so.8
 * /usr/lib/libXaw.so.6
 * /usr/lib/libXaw.so.7
 * /usr/lib/libXaw.so.8
 * /usr/lib/libXcliplist.so.1
 * /usr/lib/libXcliplist.so.1.0.0
 * /usr/lib/libXcomposite.so.1
 * /usr/lib/libXcomposite.so.1.0.0
 * /usr/lib/libXcursor.so.1
 * /usr/lib/libXcursor.so.1.0.2
 * /usr/lib/libXdamage.so.1
 * /usr/lib/libXdamage.so.1.1.0
 * /usr/lib/libXdmcp.so.6
 * /usr/lib/libXdmcp.so.6.0.0
 * /usr/lib/libXevie.so.1
 * /usr/lib/libXevie.so.1.0.0
 * /usr/lib/libXext.so.6
 * /usr/lib/libXext.so.6.4.0
 * /usr/lib/libXfixes.so.3
 * /usr/lib/libXfixes.so.3.1.0
 * /usr/lib/libXfontcache.so.1
 * /usr/lib/libXfontcache.so.1.0.0
 * /usr/lib/libXfont.so.1
 * /usr/lib/libXfont.so.1.4.1
 * /usr/lib/libXft.so.2
 * /usr/lib/libXft.so.2.1.13
 * /usr/lib/libXinerama.so.1
 * /usr/lib/libXinerama.so.1.0.0
 * /usr/lib/libXi.so.6
 * /usr/lib/libXi.so.6.1.0
 * /usr/lib/libXmu.so.6
 * /usr/lib/libXmu.so.6.2.0
 * /usr/lib/libXmuu.so.1
 * /usr/lib/libXmuu.so.1.0.0
 * /usr/lib/libXpm.so.4
 * /usr/lib/libXpm.so.4.11.0
 * /usr/lib/libXprintAppUtil.so.1
 * /usr/lib/libXprintAppUtil.so.1.0.0
 * /usr/lib/libXprintUtil.so.1
 * /usr/lib/libXprintUtil.so.1.0.0
 * /usr/lib/libXp.so.6
 * /usr/lib/libXp.so.6.2.0
 * /usr/lib/libXrandr.so.2
 * /usr/lib/libXrandr.so.2.2.0
 * /usr/lib/libXrender.so.1
 * /usr/lib/libXrender.so.1.3.0
 * /usr/lib/libXRes.so.1
 * /usr/lib/libXRes.so.1.0.0
 * /usr/lib/libXss.so.1
 * /usr/lib/libXss.so.1.0.0
 * /usr/lib/libXTrap.so.6
 * /usr/lib/libXTrap.so.6.4.0
 * /usr/lib/libXt.so.6
 * /usr/lib/libXt.so.6.0.0
 * /usr/lib/libXtst.so.6
 * /usr/lib/libXtst.so.6.1.0
 * /usr/lib/libXvMC.so.1
 * /usr/lib/libXvMC.so.1.0.0
 * /usr/lib/libXvMCW.so.1
 * /usr/lib/libXvMCW.so.1.0.0
 * /usr/lib/libXv.so.1
 * /usr/lib/libXv.so.1.0.0
 * /usr/lib/libXxf86dga.so.1
 * /usr/lib/libXxf86dga.so.1.0.0
 * /usr/lib/libXxf86misc.so.1
 * /usr/lib/libXxf86misc.so.1.1.0
 * /usr/lib/libXxf86vm.so.1
 * /usr/lib/libXxf86vm.so.1.0.0
 * /usr/lib/libgomp.so.1
 * /usr/lib/libpng14.so.14
 * /lib/libz.so.1
 * /usr/lib/libltdl.so.7
 * /usr/lib/libasound.so.2
 * /usr/lib/libao.so.2
 * /usr/lib/libFLAC.so.8
 * /usr/lib/libgsm.so.1
 * /usr/lib/libpulse.so.0
 * /usr/lib/libpulse-simple.so.0

Files actually copied. (Did not already exist on Caesar)

/usr/lib 64K -rwxr-xr-x. 1 root root     62K Feb 17 16:14 libgomp.so.1 172K -rwxr-xr-x. 1 root root    170K Feb 17 16:16 libpng14.so.14 40K -rwxr-xr-x. 1 root root     38K Feb 17 16:25 libltdl.so.7 876K -rwxr-xr-x. 1 root root    875K Feb 17 16:26 libasound.so.2 20K -rwxr-xr-x. 1 root root     19K Feb 17 16:27 libao.so.2 344K -rwxr-xr-x. 1 root root    344K Feb 17 16:30 libFLAC.so.8 48K -rwxr-xr-x. 1 root root     46K Feb 17 16:31 libgsm.so.1 288K -rwxr-xr-x. 1 root root    288K Feb 17 16:34 libpulse.so.0 16K -rwxr-xr-x. 1 root root     14K Feb 17 16:35 libpulse-simple.so.0 124K -rwxr-xr-x. 1 mcy59 cis790 122K Feb 18 17:42 libxcb.so.1.1.0 0 lrwxrwxrwx. 1 root root      24 Feb 18 17:44 libxcb.so.1 -> /usr/lib/libxcb.so.1.1.0

Files moved to /usr/lib/DELETE

/lib 76K -rwxr-xr-x. 1 root root 74K Feb 17 16:19 libz.so.1

Files moved to /lib/DELETE

command history
 * Results:
 * 501 mkdir DELETE
 * 502 pwd
 * 503 ls -slrath
 * 504 mv libgomp.so.1 DELETE/
 * 505 mv libpng14.so.14 DELETE/
 * 506 cd DELETE/
 * 507 ls
 * 508 cd ..
 * 509 pwd
 * 510 mv libltdl.so.7 DELETE/
 * 511 mv libasound.so.2 DELETE/
 * 512 mv libao.so.2 DELETE/
 * 513 mv libFLAC.so.8 DELETE/
 * 514 mv libgsm.so.1 DELETE/
 * 515 mv libpulse.so.0 DELETE/
 * 516 mv libpulse-simple.so.0 DELETE/
 * 517 mv libxcb.so.1.1.0 DELETE/
 * 518 mv libxcb.so.1  DELETE/
 * 519 ls -lsarth
 * 520 sox
 * 521 ls -slrath
 * 522 cd DELETE/
 * 523 ls
 * 524 l s-l
 * 525 ls -l
 * 526 cd /lib
 * 527 ls -lsrath
 * 528 cd udev/
 * 529 ls
 * 530 ls -slrath
 * 531 cd ..
 * 532 ls -lsrath
 * 533 pwd
 * 534  76K -rwxr-xr-x.  1 root root  74K Feb 17 16:19 libz.so.1
 * 535 pwd
 * 536 mkdir DELETE
 * 537 mv libz.so.1 DELETE/

Sox broken again.

--- rsync readme --- (unfinished) rsync backup setup readme Tom Rubino 4-21-16

Backup system using rsync and hard links

Working dir - reccomend new partition/drive/system for backups in future /mnt/main/rsyncBackups

The working dir is where backups currently reside.

Currently backed up dirs include:

/etc /usr/lib /mnt/main/bin /mnt/main/local /mnt/main/scripts


 * More system and user dirs will be added in the future.

Schedule Currenly manually run. I will update this section when schedule is determined and implemented.

Folder structure backup.0 = latest backup backup.1 = one cycle earlier than backup.0 backup.2 = two cycles earlier than backup.0 etc....

Backup folder contents This backup system takes advantage of hard linking in linux systems. Each backup folder will appear to be a full copy.

The initial backup or oldest backup will include real copies of files. All newer folders are created using hard links. These are a special type of file link that points the the same inode location and therefore the same file.

To understand how this works it is important to understand how rm (remove) file works in linux. When you rm a file Example: rm test.txt The text.txt file is not actually removed, but instead the pointer to the file is removed. This means the inode or physical location of the file remains the same, but the system thinks this space is available for use.

With hard linking you can create more than one link to a file or inode. So if your original file is text.txt and you create a hard link to the same file say text2.txt (yes you can make the name different) the file is not actually copied, but instead a second link to the same file is created using much less space (only the metadata needed to setup pointer).

As I mentioned previously if you delete the first file you are not actually deleting it, just updated the number of pointers to the file. In this case we have 2 pointers or links. So your count would change from 2 to 1 meaning the file still exists on the system.

Running a new backup. - script not yet written 2-21-16

What happens when this script runs? When a new backup is started the existing folders are first moved or renamed.

Starting with the oldest backup the folder is renamed mv backup.x to backup.x+1

example might be mv backup.17 backup.18

Each folder is renamed until backup.0 becomes backup.1 and there is no backup.0.

Backup.0 is then created by hard linking to everything in backup.1 Example: cp -al Backup.1 Backup.0

We now have Backup.0 in the latest state.

A fresh set of rsyncs are run against this dir. One for each dir being backed up. These rsyncs are set to cleanout anything not currently in the source dir.

Example: rsync -avn --delete /etc /mnt/main/rsyncBackups/Backup.0/etc

This will update Backup.0/etc to look exactly like /etc

'''This is personal log entry and an incomplete file, so ignore it... For most up to date instructions check README.txt file in rsyncBackups dir or dedicated wiki page (not yet created)'''

Created rsync page.
 * Task:

Reviewed changes to sox page.

Sox hotfix implemented in /mnt/main/local/lib/hotfix/opensuse
 * Task: 2/29.16

two dirs created usr and lib respectively to keep the same format from source


 * contents of /mnt/main/local/lib/hotfix/opensuse/usr/lib

-rwxr-xr-x. 1 root root    62K Feb 17 16:14 libgomp.so.1 -rwxr-xr-x. 1 root root   170K Feb 17 16:16 libpng14.so.14 -rwxr-xr-x. 1 root root    38K Feb 17 16:25 libltdl.so.7 -rwxr-xr-x. 1 root root   875K Feb 17 16:26 libasound.so.2 -rwxr-xr-x. 1 root root    19K Feb 17 16:27 libao.so.2 -rwxr-xr-x. 1 root root   344K Feb 17 16:30 libFLAC.so.8 -rwxr-xr-x. 1 root root    46K Feb 17 16:31 libgsm.so.1 -rwxr-xr-x. 1 root root   288K Feb 17 16:34 libpulse.so.0 -rwxr-xr-x. 1 root root    14K Feb 17 16:35 libpulse-simple.so.0 -rwxr-xr-x. 1 mcy59 cis790 122K Feb 18 17:42 libxcb.so.1.1.0 drwxr-xr-x. 3 root root   4.0K Feb 29 21:46 .. lrwxrwxrwx. 1 root root     15 Feb 29 22:05 libxcb.so.1 -> libxcb.so.1.1.0 drwxr-xr-x. 2 root root   4.0K Feb 29 22:05.

contents of /mnt/main/local/lib/hotfix/opensuse/lib -rwxr-xr-x. 1 root root 74K Feb 29 21:44 libz.so.1

0 lrwxrwxrwx. 1 root root   56 Feb 29 21:55 libgomp.so.1 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libgomp.so.1 0 lrwxrwxrwx. 1 root root   58 Feb 29 21:56 libpnh14.so.14 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libpng14.so.14 0 lrwxrwxrwx. 1 root root   58 Feb 29 21:58 libpng14.so.14 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libpng14.so.14 0 lrwxrwxrwx. 1 root root   56 Feb 29 21:59 libltdl.so.7 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libltdl.so.7 0 lrwxrwxrwx. 1 root root   58 Feb 29 22:00 libasound.so.2 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libasound.so.2 0 lrwxrwxrwx. 1 root root   54 Feb 29 22:00 libao.so.2 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libao.so.2 0 lrwxrwxrwx. 1 root root   56 Feb 29 22:00 libFLAC.so.8 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libFLAC.so.8 0 lrwxrwxrwx. 1 root root   55 Feb 29 22:01 libgsm.so.1 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libgsm.so.1 0 lrwxrwxrwx. 1 root root   57 Feb 29 22:01 libpulse.so.0 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libpulse.so.0 4.0K lrwxrwxrwx. 1 root root   64 Feb 29 22:01 libpulse-simple.so.0 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libpulse-simple.so.0 0 lrwxrwxrwx. 1 root root   55 Feb 29 22:04 libxcb.so.1 -> /mnt/main/local/lib/hotfix/opensuse/usr/lib/libxcb.so.1
 * Symbolic links added to /usr/lib

0 lrwxrwxrwx. 1 root root  49 Feb 29 21:59 libz.so.1 -> /mnt/main/local/lib/hotfix/opensuse/lib/libz.so.1
 * Symbolic links added to /lib

Sox working again. (See below)
 * Results:

EFFECT OPTIONS (effopts): effect dependent; see --help-effect [root@caesar lib]#

Update documentation tomorrow.
 * Plan:
 * Concerns:

Week Ending March 8, 2016
logged in, and setup ssh. also made sure I could log into majestix. Hardware info page still shows incorrect information about ip's.
 * Task:

Majestix is 192.168.10.5 It would appear that the missing files were not in fact missing. Based on history and information from prof it appears that I ran sox from Brutus just before logging out. This would have been a simple mistake because I was going back and forth between the two. So I ran sox from Brutus thinking I was on Caesar and it obviously ran fine with no dependency issues.
 * Task:

See snippet from history rsync -anv --ignore-existing /usr/lib/libX* root@caesar:/usr/lib/ rsync -av --ignore-existing /usr/lib/libX* root@caesar:/usr/lib/ sox who exit

Logged in to poke around at ssh stuff. Just got home from work, so I will look at more tomorrow. First thought was DNS, but that seems ok.

SSH seems to be working right now. Logged into Caesar and ssh'd in again worked fine.

Last login: Tue Mar 8 18:56:24 2016 from caesar / __ \     | /  \/ __ _  ___  ___  __ _ _ __     | |    / _` |/ _ \/ __|/ _` | '__|     | \__/\ (_| |  __/\__ \ (_| | |      \____/\__,_|\___||___/\__,_|_|                  Welcomes you!

[tsj22@caesar ~]$ who tsj22   pts/0        2016-03-08 18:55 (tom-x1.ra.unh.edu) tsj22   pts/1        2016-03-08 18:57 (caesar)

ssh works fine. majestix is up reviewed history logs and confirmed sox was run from brutus before logging out instead of Caesar. This makes sense based on what we talked about.
 * Results:


 * Plan:


 * Concerns:

Week Ending March 22, 2016
ran successful experiment 0286/002
 * Task:


 * Results:

success of course is yet to be determined as I am going to see if anyone else ran the first_5hr and if the results were the same.
 * Plan:


 * Concerns:

Week Ending March 29, 2016
Emailed Josh Young (UNH systems) about getting access to backup server. Wed 23rd. Have not heard back yet.
 * Task:

Reading up on sshd options and traffic for potential firewall issues. Prof Jonas hinted to look around in /etc just not sure what other files might come into play outside of /etc/ssh

gentrans error WARNING: "wave2feat.c", line 682: Can't find byte format in header, setting to machine's endian INFO: wave2feat.c(139): /mnt/main/Exp/0286/004/wav/sw2007A-ms98-a-0132.sph
 * Results:

not whole error

figured out how to convert to pcm with sox convert ulaw to PCM sox sw04940.sph --bits 16 --encoding signed-integer test.wav [root@caesar rsyncBackups]# sox --i test.wav

Input File    : 'test.wav' Channels      : 2 Sample Rate   : 8000 Precision     : 16-bit Duration      : 00:04:59.91 = 2399248 samples ~ 22493 CDDA sectors File Size     : 9.60M Bit Rate      : 256k Sample Encoding: 16-bit Signed Integer PCM

[root@caesar rsyncBackups]# sox --i sw04940.sph

Input File    : 'sw04940.sph' Channels      : 2 Sample Rate   : 8000 Precision     : 14-bit Duration      : 00:04:59.91 = 2399248 samples ~ 22493 CDDA sectors File Size     : 4.80M Bit Rate      : 128k Sample Encoding: 8-bit u-law


 * Sox fixed on drone servers.

Idefix EFFECT OPTIONS (effopts): effect dependent; see --help-effect [root@idefix lib]#

Asterix EFFECT OPTIONS (effopts): effect dependent; see --help-effect [root@asterix lib]#

Obelix EFFECT OPTIONS (effopts): effect dependent; see --help-effect [root@obelix lib]#

Miraculix EFFECT OPTIONS (effopts): effect dependent; see --help-effect [root@miraculix lib]#

Finish up documentation of sox dependencies. Implement sox hotfix on other servers Attempt to repair SSH.
 * Plan:

Just got home from another class. Attempt to finish up these things tomorrow after work. So still no luck with SSH issue. I put another several hours in, and can't seem to get anywhere on this. My next step would be to recreate all keys on Caesar and start from a default SSHD config.
 * Concerns:

ssh is a older version on Caesar too, so thats something.

A few things. On Brutus it looks like it was using protocol 1. All new machines seem to be set to 2.

[tsj22@obelix ~/.ssh]$ yum list installed | grep -i ssh libssh2.i686                             1.4.2-1.el6                         @anaconda-RedHatEnterpriseLinux-201409260744.i386/6.6 openssh.i686                             5.3p1-104.el6                       @anaconda-RedHatEnterpriseLinux-201409260744.i386/6.6 openssh-askpass.i686                     5.3p1-104.el6                       @anaconda-RedHatEnterpriseLinux-201409260744.i386/6.6 openssh-clients.i686                     5.3p1-104.el6                       @anaconda-RedHatEnterpriseLinux-201409260744.i386/6.6 openssh-server.i686                      5.3p1-104.el6                       @anaconda-RedHatEnterpriseLinux-201409260744.i386/6.6 [tsj22@obelix ~/.ssh]$

[root@caesar ssh]# yum list installed | grep -i ssh This system is not registered to Red Hat Subscription Management. You can use subscription-manager to register. libssh2.x86_64                          1.4.2-1.el6                           @anaconda-RedHatEnterpriseLinux-        201311111358.x86_64/6.5 openssh.x86_64                          5.3p1-94.el6                          @anaconda-RedHatEnterpriseLinux-     201311111358.x86_64/6.5 openssh-askpass.x86_64                  5.3p1-94.el6                          @anaconda-RedHatEnterpriseLinux-201311111358.x86_64/6.5 openssh-clients.x86_64                  5.3p1-94.el6                          @anaconda-RedHatEnterpriseLinux-201311111358.x86_64/6.5 openssh-server.x86_64                   5.3p1-94.el6                          @anaconda-RedHatEnterpriseLinux-201311111358.x86_64/6.5

Week Ending April 5, 2016
Team made a good plan Wed before we broke. Setup communication paths and file shares. Suggested we should all dive into researching different parameters of the various tools to see what me might want to adjust while we wait for initial train to run. Setup spreadsheet with some common sphinxtrain params and signed myself up for a few.

Researched the params that were assigned to me and updated our team documentation. It appears as though these particular settings will prove to be helpful, but we shall see.
 * Task:

Emailed Experiments group to see how I can help with wiki issues. Going to start reading up on MediaWiki API as well.
 * Task:

Heard back from Kevin. Read through his logs and copied the http.pl and https2.pl files in his home dir to mine so I can muck around with them without interrupting his work.

Want to talk to him before really attempting to run it as I'm still new to wiki API.

Did some more poking around on emacs install on majestix. Running a simple whereis emacs command gave me a place to start [root@majestix emacs]# whereis emacs emacs: /usr/bin/emacs /usr/libexec/emacs /usr/share/emacs /usr/share/man/man1/emacs.1.gz
 * Task:

Followed up with (locate -i emacs) used to find all paths that contain "emacs" insensitive. The result is approx 2178 lines.

File too large for here. Left txt file emacslist.txt in root's home folder on majestix. Emailed team

---Email---

So I poked around a bit more on majestix. and found 2178 files that are tied to emacs. Now, all these files probably are not necessary to make it run, BUT they are indeed a result of installing with yum.

You can view the list of files by taking a peak at emacslist.txt in root's home dir on majestix.

This is not a complete list of files needed for emacs, but rather all files with "emacs" (case insensitive) in the path. So no dependencies are listed here, like libotf.so.0 (which caesar is asking for when trying to run emacs)

If we truly want to mimic majestix there is a bit of work to be done. I'm not sure if we should continue down this path as it will probably be cleaner and easier to install from source, OR better yet install all the tools with YUM setup sphinx and nfs and test to verify the results are the same.

In the real world I would probably set this server up the way I like it (emacs, screen, tree whatever else I want) and then setup the sphinx tools and make sure everything works, and the results are the same. Then use this machine as the gold master image to push out to the rest.

So far still nothing from Josh Young. Majestix still down, so no progress there.
 * Results:

Heard from Josh April 5th around 11am. replied back asking for details to get connected to hyper-v server. Hope to stp in tomorrow and get things going.


 * Plan:


 * Concerns:

Week Ending April 12, 2016
Research params as outlined by team.
 * Task:

Running train 008 on Miraculix to test changes.

Researched HD pricing for rsync server. Email sent to Jonas with recommendation. So looks like a 3TB drive can be had for just over $107, whereas a 4TB drive can be had for around $130. We don't need anything super fast, buffer speed really isn't an issue as we will really be network and CPU limited anyway. Considering the first full snapshot is going to be over 800GB and a fairly small price difference I think we should go 4tb.

3TB http://www.amazon.com/Red-3TB-Hard-Disk-Drive/dp/B008JJLW4M/ref=sr_1_28?s=pc&ie=UTF8&qid=1460244985&sr=1-28&keywords=4tb+3.5

4TB http://www.amazon.com/Blue-Desktop-Hard-Disk-Drive/dp/B013HNYV8I/ref=sr_1_2?s=pc&ie=UTF8&qid=1460245731&sr=1-2&keywords=4tb+3.5 Prof ordered 4TB as it comes with a better warranty http://www.amazon.com/dp/B00D1GYO4S/ref=psdc_1254762011_t1_B00EHBERSE

008 done. scoring posted. although faster WER increased quite a bit. Documented in logs. I believe the speed increased is based on narrowing the freq ranges the system is looking for and ignoring the rest as well as other factors including different loads etc...
 * Results:

Update backup wiki info and relocate to previous attempts dir. This will probably happen tomorrow as I just got home from another class. Fun.
 * Plan:

So Jonas asked me to setup test run on Majestix, but it does not have NFS. Emailed for clarification if that is the server we will use in the end.

Cleaned up backups page, and moved clonezilla and google code info onto separate pages under previous systems.

Did a few hours of work with John on emacs dependencies. Still more to be done.
 * Concerns:

Week Ending April 19, 2016
Good plan with team. Need to get python plugins going. Reading up on that now. Hope to get some time tomorrow.
 * Task:

Caesar really slow logins...Someone must be running a train or decode on it. Keep timing out trying to connect.

Uploaded rsnapshot (rsync scheduler) to Rome. Package rsnapshot-1.3.1-12.el6.noarch.rpm

More details here. http://rsnapshot.org/faq.html


 * New files:

rsnapshot backup system in place. Currently not scheduled and only can be run manually. I will be documenting setup and info shortly.
 * Results:

ran some initial backups of /home and /root

''Current test dir is /root/.snapshots

The system is currently setup for to keep 7 snapshots. 0-6 with 0 being the latest.

Still waiting to hear about python. 4/19/16
 * Concerns:

Wiki documented for rync current config.
 * Wiki:

https://foss.unh.edu/projects/index.php/Speech:Backups

Week Ending April 26, 2016
Work on getting python going.
 * Task:

Spent a few hours until I figured out drones seem to be installed as 32bit instead of 64bit like caesar. This will cause a problem down the road if we want to share libraries and things... Emailed prof.
 * Results:

Did get the following installed.

/usr/local renamed to /usr/local-PYTHON

NFS restored so we could test a run.

Dive back in tomorrow to see about i686 stuff.
 * Plan:

Installed

devtoolset-1.1-runtime-1-13.el6.noarch.rpm

devtoolset-1.1-gcc-4.7.2-5.el6.i686

complaining about

libmpfr.so.1

Spoke with John Trimble. Pointed him toward Emacs 23.1 so he didn't need to fight updating glibc stuff. Emacs 24.5 expects many things to be updated.
 * Tools:

http://rpm.pbone.net/index.php3/stat/4/idpl/31336449/dir/centos_6/com/emacs-nox-23.1-28.el6.i686.rpm.html


 * Task:

rsync vm has been installed and is in initial setup. It was installed with Ubuntu Server 16.04 (no gui). This system has the basic requirements for what we need, ssh, and rsync.

I will update info page with more details once setup in complete.


 * Concerns:

Emailed Prof. Jonas as we do not have an available ethernet port at the moment due to system group using the other interface for IRC.

So heard back from Prof. Going to take over the second nic for backups and IRC will be figured out later.

Had to fix Miraculix because I forgot to run "ldconfig" after remounting the NFS share. Opps.

Week Ending May 3, 2016
Need to work some more on python.
 * Task:

At this point I have downloaded and attempted to manually get all dependencies in place and it appears to be a never ending maze.

So far these have been installed and there are still a ton of things needed. I think at this point I'm going to get an outside connection up and running on Majestix and attempt to run "yum install --downloadonly --downloaddir=/root/localyum"

This will allow yum to check for dependencies and download them all to the /root/localyum dir without installing them. I can then install them manually.

Helped John get most of emcas stuff squared away and tested on obelix using the /mnt/main hotfix.

He will be documenting and setting up on the other machines.

rsync server connected and communicating with Rome. I will be setting up keys and then running a small backup for testing.

rsnapshot up and running in a dev state right now. Need to figure out plan for NFS as Rome does not currently have /mnt/main mapped.
 * Results:

Helped DK get screen up and running. We actually installed from an rpm.

Tested and ran through basic usage so DK could update the wiki page with tutorial and info.


 * Plan:


 * Concerns:

Week Ending May 10, 2016
Setup hotfix for screen, tree.
 * Task:

Tree is only a single executable (all other dependencies are built into RL6)

Screen only needs two files

/etc/screen.rc

/usr/bin/screen

The dependencies are built in (for 32bit) systems like the clones. Caesar needed the 32bit libraries linked. See details on hotfix page.

Screen, Emacs, and tree are up and running on all servers with exception of Rome & Brutus (not hot fixed for anything)
 * Results:

Double checked rsnapshot is indeed running on schedule. At present it is scheduled for midnight and is backing up /mnt/main/scripts

Finish up wiki documentation for backup system. It is in place and fully functioning including schedule.
 * Task:

Did a bunch of work on the final report, and updated the rsnapshot page. Unfortunately the rsnapshot server is no able to be reached at the moment. Emailed class and JY in 124. I was hoping to get some info form the configs and screen shots for the information page tonight. Once it is back up I will get that updated.

Did some more work on the tools group Report section. Looks pretty good. Want to talk to rest of groups to decide on formatting.
 * Task:

Connection to CapBack 192.168.1.5 still down from Rome. neil said he might be able to take a look tomorrow, so hopefully it's back and I can grab the stuff I need for the wiki.
 * Concerns:

Just what type of pizza I want.
 * Concerns:

Soooooo. Today it looks like there is a networking issue.