Puppy 4.1 Alpha 2
I reported before (in the wrong thread) that when I booted without a pup_save file the desktop was grey, ROX didn't have a pinboard. Haven't had a problem until today. I booted ram only and made a new pup_save file. This time the ram only boot had a normal desktop, but the first boot with new pup_save file had no desktop. Rebooted and all is fine. This was not a problem in alpha1.
wvdialconf freezes Puppy when probing non-serial devices
When pupdial > modemprobe > wvdialconf runs to probe for modems, wvdial freezes the system when it encounters either of the ttyS_... device nodes in scanning for modems. Running wvdialconf from a console yields only:
Is there another similar tool that tests for presence of a USB modem? PCI and sound-card modems need only be detected during bootup. But serial and USB modems also need to be detected while Puppy is running, in case they get plugged in. Similarly, we may need a way to detect that a serial or USB modem is no longer plugged in or turned on.
Richard
UPDATE 6/12/08: I suspect the problem is with the LT modem setup/driver. If I install only a SmartLink PCI modem, PupDial Probe works as before: finding the SL modem and having the ttyS_ESS0 and ttyS_PCTEL device entries in the wvdialconf output showing "no device found".
But with only a Lucent PCI modem installed, those "ttyS_" devices indicate that those modems were found,. Previously, the ESS & PCTEL modem types would appear as found. Now wvdialconf just freezes the PC. If I rename the two "/dev/ttyS_" device nodes, the probe completes. Previously the probe just showed bogus information; now, even that is not tolerated by wvdialconf.
As a workaround, I will try having the (LT, ESS, PCTEL) modem scripts rename the ESS and PCTEL nodes to remove the "S", if the current modem is an LT, and name them back if the modem is ESS or PCTEL. But is there a possibility of fixing wvdialconf to avoid the hangup, when ESS or PCTEL modems are probed while the LT driver is loaded? (End of Update)
Richard
Without the ttyS_ device nodes, the scan is:ttyS0<*1>: ATQ0 V1 E1 -- ATQ0 V1 E1 -- ATQ0 V1 E1 -- nothing.
Port Scan<*1>: S1 S2 S3
This new phenomenon implies that devices ttyS... should be reserved for actual serial modems, not linmodems. Wvdialconf is advertised as being for only serial modems. Now it appears to enforce that constraint. Therefore, one way to eliminate the freezing is to use different device node names for ESS, PCTEL, and SmartLink modems, to avoid ttyS. Perhaps tty_ESS0, tty_PCTEL0, tty_SL0. What would be impacted by those changes?ttyS0<*1>: ATQ0 V1 E1 -- ATQ0 V1 E1 -- ATQ0 V1 E1 -- nothing.
Port Scan<*1>: S1 S2 S3
ttyS4<Info>: No such device or address
Port Scan<*1>: S4
Is there another similar tool that tests for presence of a USB modem? PCI and sound-card modems need only be detected during bootup. But serial and USB modems also need to be detected while Puppy is running, in case they get plugged in. Similarly, we may need a way to detect that a serial or USB modem is no longer plugged in or turned on.
Richard
UPDATE 6/12/08: I suspect the problem is with the LT modem setup/driver. If I install only a SmartLink PCI modem, PupDial Probe works as before: finding the SL modem and having the ttyS_ESS0 and ttyS_PCTEL device entries in the wvdialconf output showing "no device found".
But with only a Lucent PCI modem installed, those "ttyS_" devices indicate that those modems were found,. Previously, the ESS & PCTEL modem types would appear as found. Now wvdialconf just freezes the PC. If I rename the two "/dev/ttyS_" device nodes, the probe completes. Previously the probe just showed bogus information; now, even that is not tolerated by wvdialconf.
As a workaround, I will try having the (LT, ESS, PCTEL) modem scripts rename the ESS and PCTEL nodes to remove the "S", if the current modem is an LT, and name them back if the modem is ESS or PCTEL. But is there a possibility of fixing wvdialconf to avoid the hangup, when ESS or PCTEL modems are probed while the LT driver is loaded? (End of Update)
Richard
Last edited by rerwin on Thu 12 Jun 2008, 21:32, edited 1 time in total.
Just to Report Ethernet Jack Not Autodetected on Inspiron 4
Just to report running 4.0 (k:2.6.21.7) on a dell inspiron 4000, the onboard RJ45 did not auto detect. I'm having a hard time finding out exactly what device is, but the v90 modem, right next to it, is a 3com, so I'm assuming th e 10/100 is 3com as well.
Anyway just to let you guys know.
Good luck with 4.1 !
Anyway just to let you guys know.
Good luck with 4.1 !
- DreamsToGo
- Posts: 118
- Joined: Fri 27 Jan 2006, 08:28
- Location: Castelnaudary, south France, World Capital of Cassoulet
I'm running 4.01 alpha 2 on my Out house PC. See earlier posts for other comments. I'm writing this one because my machine consistently fails to poweroff when I select the poweroff menu item; it shows a message saying that the 'System Halted' and doesn't power off. D4 final did not have this problem.
-
- Posts: 5464
- Joined: Fri 10 Jun 2005, 05:12
- Location: Australia
Re: Just to Report Ethernet Jack Not Autodetected on Inspiro
Well a Google search for "Inspiron 4000 linux" quickly reveals - "Intel Corporation 82557 Ethernet Pro 100"jplagarde wrote:dell inspiron 4000, the onboard RJ45 did not auto detect. I'm having a hard time finding out exactly what device is
There are two competing modules (drivers) for this chipset - eepro100 and e100. According to Google, eepro100 works the best, but you should possibly try both.
Re: Just to Report Ethernet Jack Not Autodetected on Inspiro
tempestuous,
Thanks for the reply. I had tried both of those, but just to double check I did it again. I unplugged the usb ethernet I had, rebooted with the cat5 cable plugged into the jack. THen I went to "connect" and saw that the computer did not recognize anything. So then I went to "Load" and tried both the eepro100 and the e100 (both intels) and after loading it said on both occasions that no new hardware was detected.
I'm just loading modules not clicking specify that allows for parameters.
Thanks again!
Thanks for the reply. I had tried both of those, but just to double check I did it again. I unplugged the usb ethernet I had, rebooted with the cat5 cable plugged into the jack. THen I went to "connect" and saw that the computer did not recognize anything. So then I went to "Load" and tried both the eepro100 and the e100 (both intels) and after loading it said on both occasions that no new hardware was detected.
I'm just loading modules not clicking specify that allows for parameters.
Thanks again!
SmartLink USB modem detected but driver fails to load
Booting with my SL USB modem plugged in produces a bootsysinitlog error and many error messages.
In addition, when I use only the PCI SL card, the modem is recognized but the connect to my ISP talks at the ISP but gets no response from it. So, even if slusb is corrected, it may still not work fully; but I need the fix in order to verify that.
Richard
FATAL: Error inserting slusb (/lib/modules/2.6.25.4/slmodem/slusb.ko): Unknown symbol in module, or unknown parameter (see dmesg)
Module slusb does not appear in lsmod output.slusb: Unknown symbol usb_alloc_urb
slusb: Unknown symbol usb_free_urb
slusb: Unknown symbol usb_register_driver
slusb: Unknown symbol usb_clear_halt
slusb: Unknown symbol usb_submit_urb
slusb: Unknown symbol usb_control_msg
slusb: Unknown symbol usb_set_interface
slusb: Unknown symbol usb_deregister
slusb: Unknown symbol usb_unlink_urb
In addition, when I use only the PCI SL card, the modem is recognized but the connect to my ISP talks at the ISP but gets no response from it. So, even if slusb is corrected, it may still not work fully; but I need the fix in order to verify that.
Richard
Modem not responsing
Barry,
I can't dial up with my serial USRobotics modem and pupdial. It says modem not responding. It works fine in 4.0.0.
My USB floppy's vendor and model have already been listed by someone on your blog. I tried wakepup2-20070919.pet and it doesn't find the CD drive on my new PC. Does it work with SATA drives. My CD drive might be SATA - the HDD is.
Could you put up a shield for the Pmount window before it's removed for refresh. When resources are low things don't happen so quickly and there's a danger of me clicking on an underlying window. It could be like the orange one that comes up after clicking refresh (and should come up after USB RAM insertion IMHO) but it would need to have the same size and position.
Can I thank you and all the developers for Puppy Linux.
I can't dial up with my serial USRobotics modem and pupdial. It says modem not responding. It works fine in 4.0.0.
My USB floppy's vendor and model have already been listed by someone on your blog. I tried wakepup2-20070919.pet and it doesn't find the CD drive on my new PC. Does it work with SATA drives. My CD drive might be SATA - the HDD is.
Could you put up a shield for the Pmount window before it's removed for refresh. When resources are low things don't happen so quickly and there's a danger of me clicking on an underlying window. It could be like the orange one that comes up after clicking refresh (and should come up after USB RAM insertion IMHO) but it would need to have the same size and position.
Can I thank you and all the developers for Puppy Linux.
Pmount mod
I've been wanting to comment on Pmount and my idea may just solve this situation.zygo wrote:Could you put up a shield for the Pmount window before it's removed for refresh. When resources are low things don't happen so quickly and there's a danger of me clicking on an underlying window. It could be like the orange one that comes up after clicking refresh (and should come up after USB RAM insertion IMHO) but it would need to have the same size and position.
When 'REFRESH' is clicked in Pmount, I would like the Pmount window to remain on screen with the orange 'detecting drives' window superimposed on it. The Pmount window should just refresh quickly when the orange window closes. This would be less disturbing and also avoid the problem zygo mentions.
Methinks Raspberry Pi were ideal for runnin' Puppy Linux
-
- Posts: 405
- Joined: Mon 01 Jan 2007, 20:24
- Location: Hayslope, near Middlemarch, Midlands, England
Count me in too.boscobearbank wrote:Count me in that < 0.1%. On the other hand, I have yet to determine a solution to the drive mounting/display problem that I consider ideal, I'll take whatever Barry's giving.Sage wrote:The number of folks using more than three or four partitions is probably limited to <0.1% of punters reading this Forum?
Like many who dual-boot with The Devil's Own OS I have 3 partions for 'doze (Boot, Main and Recovery) plus a fourth for Puppy which is sub-divided into more - currently 3. Just how many varies depending on what I'm trying. And then I plug in one or two USB devices.........
So I like the Hotpup layout with an icon for each partition/device. But then I have a 17" widescreen laptop so there's plenty of room.
However; with the increasing popularity of the EeePC and similar kit with a small screen I can see that Barry's compromise is rational. There just wouldn't be room for all my drives on a little screen if each has an icon of its own. So I favour an icon per physical device which opens either Rox or Pmount (or MUT, of course) whichever is appropriate.
[color=darkblue][b][size=150]Nick[/size][/b][/color]
Ummm, I think the first half makes some sense, but I don't like the second half. If a click of an icon mounts the partition, then a click of an icon (of a mounted partition) should unmount it. It's just wrong to have two different applications toggle the mount state. Also I don't like being nagged constantly when I close a Rox window, I just want the dang thing to go away.It should work like in HotPup where the drive is mounted and a ROX window opens. When the window is closed, ask if the partition should be unmounted - nice and simple!
Are we making Pmount/MUT pointless? A shame, I like them.
I think Sage makes some sense. Let the default be friendly to Windows refugees (within reason!). Let configurations options exist to please the linux hotshots or those with big systems and numerous partitions.
Alternative event manager drive icons
I recently suggested to Barry that he reconsider his choice of event manager device icons. To my eye, something more subtle is in order. Dougal's use of a colored dot for instance, is both effective and "smart".
I'm definitely no graphic artist, but I took a few minutes to design three alternatives for the hard drive. It takes Dougal's idea a bit further by employing shape and color. Please comment if you have any feelings one way or the other. Feel free to borrow and improve.
BTW, it occurs to me now that the "unmounted" icon needs no symbol.
I'm definitely no graphic artist, but I took a few minutes to design three alternatives for the hard drive. It takes Dougal's idea a bit further by employing shape and color. Please comment if you have any feelings one way or the other. Feel free to borrow and improve.
BTW, it occurs to me now that the "unmounted" icon needs no symbol.
- Attachments
-
- drive48_mounted.png
- (3.95 KiB) Downloaded 2611 times
-
- drive48_unmountable.png
- (3.89 KiB) Downloaded 2499 times
-
- drive48_umounted.png
- (3.77 KiB) Downloaded 2514 times
- BarryK
- Puppy Master
- Posts: 9392
- Joined: Mon 09 May 2005, 09:23
- Location: Perth, Western Australia
- Contact:
You'll like what I've done in the latest incarnation of the desktop icons management. It's fully configurable for the Windows-refugees and for Linux hotshots. My system does not give ROX-Filer any responsibility for mounting or unmounting a partition -- to unmount, either you use Pmount or you right-click on a desktop icon and choose 'unmount' from the menu.PaulBx1 wrote:Ummm, I think the first half makes some sense, but I don't like the second half. If a click of an icon mounts the partition, then a click of an icon (of a mounted partition) should unmount it. It's just wrong to have two different applications toggle the mount state. Also I don't like being nagged constantly when I close a Rox window, I just want the dang thing to go away.It should work like in HotPup where the drive is mounted and a ROX window opens. When the window is closed, ask if the partition should be unmounted - nice and simple!
Are we making Pmount/MUT pointless? A shame, I like them.
I think Sage makes some sense. Let the default be friendly to Windows refugees (within reason!). Let configurations options exist to please the linux hotshots or those with big systems and numerous partitions.
Of course, noone is going to appreciate it and suggest improvements until they can get their hands on it. Hopefully 4.1alpha3 isn't too far away!
[url]https://bkhome.org/news/[/url]
triangle
regarding the mounted, unmountable icons ..
I'm colour blind to a small degree - yellows and greens generally all look the same to me (and they often get used for different buttons, indicators)
Wouldn't a red triangle be more of a warning ? and more immediately obvious by both different shape and colour/contrast ?
Though an alternative would be a yellow/green triangle positioned pointing to the right > like a play button, suggesting it's playing/in use ? and either a red square or circle suggesting a warning and stop ? The square being more standard with audio/video/media players and a circle being more like a traffic sign stop (well, part of it..) ? ...
Cheers, Adam.
I'm colour blind to a small degree - yellows and greens generally all look the same to me (and they often get used for different buttons, indicators)
Wouldn't a red triangle be more of a warning ? and more immediately obvious by both different shape and colour/contrast ?
Though an alternative would be a yellow/green triangle positioned pointing to the right > like a play button, suggesting it's playing/in use ? and either a red square or circle suggesting a warning and stop ? The square being more standard with audio/video/media players and a circle being more like a traffic sign stop (well, part of it..) ? ...
Cheers, Adam.
Well, using different shapes takes care of the color-blind problem. The question seems to be, what shapes?
It's a bit strange to mix traffic sign shapes with computer stuff, also there is the question of how universal are the sign shapes? Stop is a red octagon in the US. A yellow triangle is warning but it's inverted and more toward orange than dogone's example.
I don't have much of an opinion on the matter other than to say I like the colored dot idea generally. Sounds like you have a good plan Barry. I did like past examples warning to not pull mounted drives as well, but is that too obvious? Does everyone know that?
We'll use a different icon for flash drives I assume.
One could spend eons picking through different ways of handling icons, or arguing esthetics.
It's a bit strange to mix traffic sign shapes with computer stuff, also there is the question of how universal are the sign shapes? Stop is a red octagon in the US. A yellow triangle is warning but it's inverted and more toward orange than dogone's example.
I don't have much of an opinion on the matter other than to say I like the colored dot idea generally. Sounds like you have a good plan Barry. I did like past examples warning to not pull mounted drives as well, but is that too obvious? Does everyone know that?
We'll use a different icon for flash drives I assume.
One could spend eons picking through different ways of handling icons, or arguing esthetics.
Instead of a yellow triangle, perhaps one of these red circles with a line through it placed in the upper left hand corner. Much smaller of course, like the size of the green dot.
Isn't this a pretty much universal sign?
Isn't this a pretty much universal sign?
- Attachments
-
- NOSYMBL.gif
- (14.88 KiB) Downloaded 6740 times
Long past time to completely abandon GXine. There are far too many issues on far too much kit. It doesn't work properly on most kit older than a couple of years nor on my new lcd TV. The fora, including this one, is awash with cries of despair about this useless app. So far, I found that Xine and Goggle, (MU introduced this in his mini) are streets ahead. If this involves different compilers & w.h.y., so be it. Watching DVDs and all that video stuff has become a core function demanded by users today, displacing some of the interest in digital photography of last year. Not my bag, but it's importance cannot be denied. It's as important as blasting large holes in the silly nonsense the Yanks try to propagate over software patents, DRM, CSS and all their other greedy leveraging capitalistic monopolistic practices. Let's hope the departure of the idiot son brings enlightenment.