Lost At ‘repository’ Entry Installing CentOS7

Home » CentOS » Lost At ‘repository’ Entry Installing CentOS7
CentOS 13 Comments


while attempting to install c7, i got lost at ‘repository’ entry.

i canceled, loaded CentOS.org, looked for help for installing c7, but did not find.

i know, i did not look in right place.

if such has been posted, i missed.

so what/where is ‘the right place’?

much appreciate some help.

13 thoughts on - Lost At ‘repository’ Entry Installing CentOS7

  • Sounds like he is doing a network install, and is looking for the network path that must be supplied in order to do the install. If he doesn’t have a local repository, then he has to supply the first part of the path (e.g. http://…./xyz/ ) and he has to stop at the directory level above …/7/
    or some such. I have to look it up, and don’t remember where I have it stashed, since I haven’t done a network install in many moons. I have sometimes resorted to putting in a partial path, and then looking at the error message to see if I have to put more or less path into the box. I
    have always thought that this is one of the worst documented spots in the installation documentation. I suppose RH thinks only people with local, custom repos are using network install, and those people do it so many times that they don’t need to be reminded what to put in the box. Ted Miller Elkhart, IN, USA

  • re-sending to list. other attempt got addressed wrong.

    excuse my prior terseness.

    i pulled;


    “burned” it to a usb memory stick with;

    su -c “dd if

  • correct.

    i can agree to that. 2 days of searching both CentOS and rh sites provide nothing.

    another 2 days of web searching with same nothing.

    so i ‘bit the bullet’ and pulled the live iso.

    new problem came up in that in attempting to install to sdb3 partition, i noted that my sdc partition was divided. partitions where i had sdc1, sdc2, sdc3 and the extended partition was show as sdd5, sdd6, sdd7.

    knowing that i did not have a 4th hdd, i quit install again. stayed off installing for a day, went back again to look at what i was seeing.

    sure enough, still there, quit again, pulled the dvd iso, ran install from it, ‘God was offering little green apples’ again.

    decided to beat the odds and reconnect the drives, ie, swap the cables of 1st and 3rd drive. that is when the ‘wtf’ rang out.

    this is an hp box and in infinite wisdom of hp, there are 4 sata connectors labeled as;

    connectors to sata drives w/ dev label
    [sata0] [sata5] [hdd1] [dvd ] [sda] [sr1]
    [sata1] [sata4] [hdd2] [hdd3] [sdb] [sdc]

    i am not asking why, but i sure am curious as to why with hdd3, primary partitions show as sdc1, sdc2, sdc3, and extended show as sdd5, sdd6, sdd7.

    ‘any who’, because sata0 had oos 95 and no longer to be used, i wiped, partitioned, and formatted sata0 for install of CentOS 7.

    hopefully, this evening my ulcers will be calmed down enough so that any new surprises will not cause an ulcer to rupture.

    there may be some who believe that the new install process gui is
    ‘super fine’. there are other names i could call it, but as long as the CentOS devs continue as they otherwise have, i will maintain and contain my other thoughts. ;-)

    “God willing and the creeks don’t rise”, tomorrow i will be enjoying CentOS 7.

    as a side question, what is a good way to get a list of _all_ packages currently installed to CentOS 6.6 so that i can install them to new install? [note positive thinking] or, an easy script or what ever?

    also, in closing, my most _great-full_thanks_ to Tim and Ted for not being too evolved with “Another Fedora decision” and taking time to reply to this thread.

  • If you are new to CentOS, it really is best to download the DVD. The net-install is really designed for people who put up their own server for doing 200 installs in one building. The ipxe.iso that you tried absolutely requires a special server on your local network. Yes, you can do a single install with the net install, but it should be your 10th or 20th, not your first.

    I hope you can download the DVD and enjoy CentOS 7. “Get CentOS 7 Now” to
    “DVD ISO” to the list, and pick any link off of the list. The DVD will be quite self-explanatory. The only caveat is to make sure you go into the link for your network card and configure it. Otherwise CentOS will start up with the network card turned off.

    Ted Miller Elkhart, IN, USA

  • < <>>

    expecting possible problems, i pulled network, live, and dvd iso file at same time. all passed sha256sum check.

    when i ‘dd’ dvd iso to a usb stick, it ran ok, but when trying to define partition to install to, i keep getting message that drive was not defined again, even after having wiped sda before running c7 install, so i exited install. maybe i should not have pre sliced drive.

    my thinking shifted from a straight install of c7 to using a fresh and updated c6 install and run “yum upgrade CentOS-7.0-1406”.

    next i installed c6.4 to sda as sda1= /boot, sda3= swap, sda3= /, sda5= /home.

    not clearly thinking, when setting time zone, i unchecked hardware clock as utc. which i realized after installing c6.4 and after running “yum update”.

    there seems to be something about yum and hw clock and utc, because yum changed hw clock to utc, which i noted when i closed terminal i had run yum in. i have had this happen before.

    so an ‘in between’ question, how do i go about changing /etc/localtime so that i can reboot, change bios clock to utc and have desktop show correct utc-6 time with bios set to utc time?

  • Could you not see any drives? Or that you there wasn’t space to install on that drive?

    Upgrading to CentOS7 from CentOS6 isn’t as sumple as ‘yum upgrade CentOS 7.0-1406’. You need to follow the instructions here:


    Why CentOS 6.4? 6.6 is the latest release. There are a bunch of security holes in 6.4’s installation media.

    Look in /etc/sysconfig/clock to tell the system that your clock is UTC. The GUI tool ‘system-config-date’ (in a package with the same name) is a graphical tool for setting date/timezone settings.

  • < <>>

    yes, c7 install shows sda, sdb and sdc.

    as stated above “fresh and updated c6 install” which brought it up to c6.6.

    pulled page for off-line ref.

    the install was updated. see “as stated above” above.

    /etc/sysconfig/clock shows ZONE=”Etc/GMT-6″.

    running ‘system-config-date’ from cli, and setting hardware clock to UTC and system clock to CST, several times, ‘hwclock’ kept showing clock to be CST. so, i ran ‘system-config-date’ one more time and selected UTC for both and set clock to UTC time. weird, but that set bios clock to UTC and i was able to open ‘System Settings’
    window, select ‘Date & Time’, and set system time to correct time using chicago as time zone. too bad it does not have CST in settings because i live in memphis, tn. ;-)

    shame all that could not bet set correctly using ‘hwclock’.

    thank you for replying.

    now for some ‘head rest’, then some reading of CentOS upgrade tool before i go back to attempt install.

  • < <<>>>

    so much for all that.

    i just thought it was working. :-(

    only way i can get system clock to show correct CST is to set bios clock to CST.

  • I suggest reading the man page for ‘hwclock’. Namely, the –utc option. If you don’t tell your system that the BIOS is stored as UTC, then it will assume it’s local time.

  • i have run ‘man hwclock’ so many times that the a, c, h, k, l, m, n, o keys have gray lettering and the rest are black. so i now have one terminal with ‘man hwclock’ left open. :-D

    [geo@boxen ~]$ su Password:
    [root@boxen geo]# hwclock –utc Sat 07 Feb 2015 05:28:00 PM CST -0.219402 seconds
    [root@boxen geo]# hwclock –utc –date=”2/07/15 23:30:00″
    Sat 07 Feb 2015 05:29:35 PM CST -0.390922 seconds
    [root@boxen geo]# hwclock –set –utc –date=”2/07/15 23:30:00″
    [root@boxen geo]# hwclock Sat 07 Feb 2015 11:30:11 PM CST -0.828490 seconds
    [root@boxen geo]# hwclock –utc –set –date=”2/07/15 23:30:00″
    [root@boxen geo]# hwclock Sat 07 Feb 2015 11:30:04 PM CST -0.672246 seconds
    [root@boxen geo]# hwclock –set –date=”2/07/15 23:30:00″ –utc
    [root@boxen geo]# hwclock Sat 07 Feb 2015 11:30:08 PM CST -0.734744 seconds
    [root@boxen geo]# hwclock –set –utc –date=”2/07/15 23:30:00″
    [root@boxen geo]# hwclock Sat 07 Feb 2015 11:30:03 PM CST -0.531615 seconds
    [root@boxen geo]# hwclock –utc –set –date=”2/07/15 23:30:00″
    [root@boxen geo]# hwclock Sat 07 Feb 2015 11:30:03 PM CST -0.422288 seconds
    [root@boxen geo]#

    in fact, i have run combination’s so many times and of various ways, only difference i can think of is changing the sequence, which still does no good.

    or, i have run it so many different ways that my brain is failing to think of right combination.

    running “system-config-date”, the only way i can get bios clock to show UTC is by setting both system and zone to UTC, which makes system UTC.

    /etc/adjtime is still;

    6.736246 1423375017 0.000000

    [root@boxen etc]# grep UTC *
    grep: extlinux.conf: No such file or directory
    Binary file localtime matches
    [root@boxen etc]# grep GMT *
    grep: extlinux.conf: No such file or directory
    Binary file localtime-org matches
    protocols:gmtp 100 GMTP # GMTP
    services:n1-rmgmt 4447/tcp # N1-RMGMT
    services:n1-rmgmt 4447/udp # N1-RMGMT
    [root@boxen etc]# grep CST *
    grep: extlinux.conf: No such file or directory
    Binary file localtime-0001 matches
    services:subntbcst_tftp 247/tcp # SUBNTBCST_TFTP
    services:subntbcst_tftp 247/udp # SUBNTBCST_TFTP
    services:rcst 3467/tcp # RCST
    services:rcst 3467/udp # RCST
    services:cst-port 3742/tcp # CST – Configuration & Service \
    services:cst-port 3742/udp # CST – Configuration & Service \
    [root@boxen etc]#

    that is about it for this ‘chemo brain’ to come up with.

  • < <>>

    and that is where problem is/was.

    having looked at files that related to problem, i realized that there was one that i had not changed, but did note it had a recent time stamp.

    what changed it, i do not know, but it was time stamped during when i was playing with hwclock and system-config-date.

    being that it was showing UTC, i figured i had nothing to lose, so i changed to CST and rebooted.

    during bios, i broke to it, set bios clock to UTC again, rebooted.

    that did it. time in panel clock is now correct CST and UTC is also correct.

    now my ‘chemo brain’ can rest for a while. ;-)

    my thanks to all for responding.