summaryrefslogtreecommitdiffstats
path: root/isys
Commit message (Collapse)AuthorAgeFilesLines
...
* * isys/isys.c (doDhcpNetDevice): Rename to be dhcp instead ofJeremy Katz2006-09-123-10/+14
| | | | | | | | pump. Prefer DHCPv4 for now * isys/isys.py (dhcpNetDevice): Rename pumpNetDevice to this, leave pumpNetDevice stub * isys/Makefile (nltest): Fix build of test program
* * isys/nl.c: Only collect IPv4 and IPv6 addresses for IFF_RUNNINGDavid Cantrell2006-09-111-11/+18
| | | | interfaces.
* Skip attached devices without media present (like CF readers, etc.) (#205465).Chris Lumens2006-09-111-0/+5
|
* * isys/isys.py (inet_convertPrefixToNetmask): Take IPv4 CIDR prefixDavid Cantrell2006-09-061-0/+7
| | | | | | | | | | | | | | | | | and convert to dotted-quad netmask notation. * iw/network_gui.py (NetworkWindow.editDevices): Allow user to enter CIDR prefix or dotted-quad netmask for IPv4 settings, validate the IPv4 and IPv6 prefix and make sure they fall within the required range, if user enters IPv4 CIDR prefix then convert to dotted-quad netmask for the sysconfig ifcfg-ethX file. * iw/network_gui.py (NetworkWindow.editDevices): Call get_text() to get the value of the ipv6prefix box rather than using an instance of a gtk.Entry() object. * iw/network_gui.py (NetworkWindow.getScreen): Set hostname, gateway, and DNS entry boxes to be 41 characters wide.
* * isys/isys.py: Remove inet_aton() and inet_ntoa(). Replaced callsDavid Cantrell2006-09-061-15/+20
| | | | | | | | | | | | | | | | | | | to these functions with inet_pton() and inet_ntop(). * isys/isys.py: Added inet_convertNetmaskToPrefix() utility function to convert dotted-quad IPv4 netmasks to CIDR prefix values. * iw/network_gui.py (createIPRepr): Renamed to createIPV4Repr() and have it write out ip/prefix rather than ip/netmask. * iw/network_gui.py (createIPV4Repr): Added to handle IPv6 write outs to the tree store. * iw/network_gui.py (NetworkWindow.editDevice): Add IPv6 info to the tree store. * iw/network_gui.py (NetworkWindow.setupDevices): Initialize tree store with IPv4 and IPv6 information.
* - move device blacklisting from hardDriveDict into driveDict, so we do itPeter Jones2006-08-251-22/+37
| | | | | less often (but in more codepaths :/). Also add filtering of DGC LUNZ devices, which don't ever represent addressable storage.
* 2006-08-24 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-08-242-4/+11
| | | | | | | | | | * fsset.py (FileSystemType.mount): Reset file context on mountpoints (#202525) * packages.py (setFileCons): Use new parameter to resetFileContext * isys/isys.py (resetFileContext): Allow passing a rootpath * isys/isys.c (doResetFileContext): Likewise.
* - add a simple audit daemonPeter Jones2006-08-225-1/+174
|
* 2006-08-22 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-08-221-0/+3
| | | | * isys/isys.py (getDasdState): Fix unformatted dasds again (#203260)
* - oops, klassArg crap and dm is still busted when klassArg == "cdrom".Peter Jones2006-08-211-3/+3
| | | | This fixes that.
* - don't call isys.flushDriveDict() from makeDriveDeviceNodes(), as itPeter Jones2006-08-181-14/+22
| | | | screws up dm devices in our cache
* Revert logMessage calls (#201707).Chris Lumens2006-08-081-5/+4
|
* Don't initialize method to a valid selection, since that causes the methodChris Lumens2006-08-041-3/+3
| | | | | selection screen to never appear. This may re-break 201108, but this stuff is too tricky to debug without new CDs. Let's see what happens next time.
* Log why mount failed.Chris Lumens2006-08-031-4/+6
|
* - fix mknodPeter Jones2006-07-311-1/+1
|
* Whitespace changes only. Nothing else.David Cantrell2006-07-311-80/+81
|
* * isys/dns.c: Removed code that's not in use anymore.David Cantrell2006-07-281-32/+0
| | | | * loader2/ftp.c: Notes to myself.
* * isys/dns.c, isys/dns.h: Revert to previous version (DNS problems).David Cantrell2006-07-283-28/+173
| | | | * isys/nfsmount.c: Revert to previous version (DNS problems).
* - 4 space indentation, not 3.Peter Jones2006-07-271-356/+356
|
* * loader2/urls.c (urlinstStartTransfer): Determine address familyDavid Cantrell2006-07-263-174/+28
| | | | | | | | | | | | | | | | | | and pass that to the ftpOpen() call. * urlinstall.py (UrlInstallMethod.__init__): Use urlparse. * isys/dns.c, isys/dns.h: Updated for IPv6 support. * loader2/Makefile: Cleanups. * loader2/ftp.c, loader2/ftp.h: Updated for IPv6 support. * loader2/loader.c: Whitespace consistency. * loader2/method.c: Whitespace consistency. * loader2/urls.c: Updated for IPv6 support.
* Removed isys_calcNetmask and isys_calcNS.Chris Lumens2006-07-131-24/+0
|
* * isys/isys.c: Added back the doConfigNetDevice() andDavid Cantrell2006-07-132-0/+104
| | | | | | | | | | | doPumpNetDevice() functions, but rewritten to use libdhcp instead of pump. Since the UI has not changed for IPv6, these two functions don't really do IPv6 stuff yet. * isys/Makefile: Link and compile with libdhcp. NOTE: This requires libdhcp-1.9 (which I haven't released yet) because I moved the IP_STRLEN macro to ip_addr.h instead of having it in anaconda.
* Don't guess the gateway for consistency with loader's UI (#197578).Chris Lumens2006-07-121-8/+0
|
* Hooray for 4th grade C programming class!David Cantrell2006-06-274-10/+71
| | | | I had an idea that we could make loader smaller if we didn't call toupper() or string functions. Well, we use string functions all over the place, but I did remove the toupper() calls I used. Yeah, not exciting. Believe it or not, this was involved in a large tracking-down of double free()'s yesterday.
* - get rid of diet.Peter Jones2006-06-266-47/+5
|
* Instead of detecting whether a device is removable or not, detect whetherChris Lumens2006-06-222-118/+11
| | | | | there is media present in devices where it may be removed. This should finish up getting USB installation working.
* 2006-06-21 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-06-211-2/+1
| | | | | | | | * loader2/net.c (setupNetworkDeviceConfig): Remove s390 hack * network.py (sanityCheckIPString): Remove s390 hack * isys/isys.py (sync): Remove s390 hack
* 2006-06-21 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-06-211-11/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * iw/iscsi_gui.py (iscsiWindow.getNext): Don't need to do the startup here now. * isys/isys.py (compareDrives): Sort xvd devices first since they're the "bootable" ones for Xen (driveIsIscsi): Determine if a drive is iscsi or not. Kind of ugly. * zfcp.py (ZFCP.write): Copy zfcp.conf here. * yuminstall.py (YumBackend.doPreInstall): Write out network, zfcp and iscsi info prior to the install starting. * partitioning.py (partitionObjectsInitialize): Flush drive dict on initialization, set up iscsi devices here, make device nodes for drives * kickstart.py (AnacondaKSHandlers.doIscsi): Initial iscsi kickstart support. This syntax *will* change * iscsi.py (iscsi.action): Set nodes to start up automatically on boot in the db (iscsi.startup): Give a popup while we're waiting on iscsi devs to initialize (iscsi.write): Write out iscsi config * fsset.py: Add various bits so that we can set that a device should be marked as _netdev in the fstab and use it appropriately for iscsi.
* just the sync bitJeremy Katz2006-06-161-22/+11
|
* 2006-06-15 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-06-161-12/+24
| | | | * isys.py (sync): sync seems to cause an oops on s390 (#195630)
* 2006-06-15 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-06-151-1/+1
| | | | * isys/isys.py (inet_ntoa): Fix on 64bit machines
* 2006-06-10 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-06-101-0/+1
| | | | * isys/isys.py (readReiserFSLabel_int): Close leaky fd
* HERE COMES YOUR WATER! ....David Cantrell2006-06-072-81/+1
| | | | | | Ok, so what exactly are we looking at here? A slapped-together pile of stuff that uses libdhcp instead of pump. I call it a work in progress. Right now we have support for regular IPv4-based installs. IPv6 is sort-of halfway there. I'm still wedging it in to place (i.e., dodging sigsegvs). I wanted to get an IPv4-libdhcp-move committed now so that it's in CVS. The pump-compatible API in libdhcp is mostly useless, so I'm moving things from using the core-dumping pump API functions to just using the libdhcp stuff directly. I had to back out a lot of IPv6 stuff I did in the interim. Please avoid making huge changes to the loader2 network code because I'm still tearing through that.
* Fix on 64-bit machines (#193609).Chris Lumens2006-05-311-1/+1
|
* * isys/Makefile: Remove getipaddr.o reference.David Cantrell2006-05-264-58/+2
| | | | | | | | | * isys/getipaddr.c: Removed. * isys/isys.c (doGetIPAddress): Call netlink_interfaces_ip2str() instead of getIPAddr(). * isys/net.h: Removed getIPAddr() prototype.
* * isys/nl.c (netlink_format_ip_addr): Rewrote to malloc for spaceDavid Cantrell2006-05-262-17/+86
| | | | | | | | | | | | | | | | | | | in the function and handle IPv4 and IPv6 address lengths more gracefully. * isys/nl.c (netlink_interfaces_ip2str): Added function to return the IP address of the specified device name. If you have an IPv4 address for the device, that's returned. If you only have an IPv6 address, that's returned. The string returned is the human-readable IP addr. * isys/nl.c (_netlink_interfaces_elem_find): Renamed the _netlink_interfaces_elem_mac2str function since it's a generic intf search routine. * isys/nl.c (netlink_init_interfaces_list): If there is no IPv6 address, memset intfinfo->ip6_addr to all zeros. * isys/nl.h: Prototype updates.
* * isys/Makefile: Remove getmacaddr references. Add glib CFLAGS andDavid Cantrell2006-05-265-98/+12
| | | | | | | | | | | | | | | | | | | | | | LDFLAGS. * isys/getmacaddr.c: Removed. * isys/getmacaddr.h: Removed. * isys/isys.c: Call netlink_interfaces_mac2str() in doGetMacAddress. * isys/nl.c (netlink_format_mac_addr): Malloc space for buf if it's NULL when we enter the function. * isys/nl.c (netlink_interfaces_mac2str): Return buffer malloc is handled by netlink_format_mac_addr). * loader2/Makefile: Add glib CFLAGS and LDFLAGS. * loader2/net.c: Call netlink_format_mac_addr() and netlink_interfaces_mac2str(). * loader2/urlinstall.c: Call netlink_interfaces_mac2str() instead of getMacAddr().
* * isys/nl.h: Include glib.h and netinet/in.h.David Cantrell2006-05-251-0/+3
|
* * isys/nl.h: Prototypes for netlink_interfaces_list_free() andDavid Cantrell2006-05-252-1/+99
| | | | | | | | | netlink_interfaces_mac2str()...and their helper functions. * isys/nl.c: Several things... In netlink_format_mac_addr, make the formatted MAC address uppercase. Added netlink_interfaces_mac2str which searches the interface list and hands you the human-readable MAC address for the specified interface. Added netlink_interfaces_list_free to free the interfaces list and associated elements.
* * isys/nl.c: Skip Netlink messages with void ARP headers or zero-lengthDavid Cantrell2006-05-241-0/+7
| | | | ARP headers.
* All hail the hypnotoad!David Cantrell2006-05-243-2/+408
| | | | | | | | Yeah, so we have some netlink_* functions now. You can build nltest in the isys directory and it'll dump info on your interfaces including the index number, the name, the MAC address, the IPv4 address, and the IPv6 address. Next up is rewriting isys functions to use the netlink_* functions.
* - Fix keyboard loading hellPeter Jones2006-05-101-0/+5
|
* * isys/isys.py (driveIsRemovable): Split module checking out to itsPeter Jones2006-05-091-22/+31
| | | | | | | | | | own functions. Allow usb-storage installs. * partitions.py: warn if you're using firewire or usb * loader2/modules.c (loadModule): Delay on reloading usb-storage, as it takes undefined time as well. Without this /tmp/scsidisks is wrong. * loader2/modules.c (writeModulesConf): don't exclude usb-storage or sbp2.
* - actually don't warn about usb there, because that codepath isn't dependentPeter Jones2006-05-091-4/+1
| | | | on actually having the device in use.
* - allow usb-storage installsPeter Jones2006-05-091-1/+4
|
* Removed nfs_mountversion.h since that's generated by the Makefile. IgnoreDavid Cantrell2006-05-092-1/+1
| | | | it for other cvs ops.
* s390 build fix.Chris Lumens2006-05-081-1/+1
|
* 2006-05-04 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-05-041-0/+3
| | | | * isys/isys.c: Fix the build again
* 2006-05-03 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-05-031-3/+0
| | | | | * isys/isys.c: Don't include kernel fs headers -- the e2fsprogs ones have what we need.
* 2006-05-01 Jeremy Katz <katzj@redhat.com>Jeremy Katz2006-05-021-1/+6
| | | | * isys/dasd.c: Fix the s390 build for real with a temporary hack
generated by cgit (git 2.34.1) at 2025-09-30 16:34:34 +0000