January 10, 20179 yr Peter the file is almost 3GB. I am using a file downloader program just in case the connection breaks and I have to start again. At least it then starts where it stopped. The download time for me is about 2 hours. You may like to try the same if you are not already doing so. It took me 8 hours to upload Sent from my SM-G920F using Tapatalk
January 10, 20179 yr 3GB, compressed!? Good grief. I suppose it depends on what you put in there. When I still messed around with my own raspbian based ideas I got it down to 500MB compressed, bare bones, X (the Gui) not installed. Venus is 65MB compressed, unpacks to about 750MB (which includes a full 275MB of just zeroes, an empty partition for future updates). So you have some work to do :-)
January 10, 20179 yr Downloaded with Uget on Ubuntu in 1hr 57mins on a 4Mb ADSL line which is maximum speed of Telkom exchange in our village. Will be loading it on Rpi2 instead of RPi3 which is currently running. It will be interesting to see if there is much of a difference in performance.
January 10, 20179 yr I now have the img on my hard disk. But i cant "burn" the SD card. Command "Sudo dd" returns "Operation not supported" and the Mac OS Diskutility does not have such an option. Just copy all files to the SD card? But it would't be bootable, no?
January 10, 20179 yr 1 hour ago, PeterGutti said: Operation not supported Perhaps the sdcard automounts. That always happens on my system. You have to unmount it first and then attempt to write to it. That error is fairly generic, it's what happens when a syscall returns ENOTSUP. So it could be anything, but you could try and trace it with strace :-) You did: sudo dd if=imagefile bs=1M of=/dev/mmcblk0 right?
January 10, 20179 yr 8 hours ago, plonkster said: 3GB, compressed!? Good grief. I suppose it depends on what you put in there. When I still messed around with my own raspbian based ideas I got it down to 500MB compressed, bare bones, X (the Gui) not installed. Venus is 65MB compressed, unpacks to about 750MB (which includes a full 275MB of just zeroes, an empty partition for future updates). So you have some work to do :-) 2017-01-11-raspbian-jessie.zip = 1.42gb. I installed all the requirements and the app is about 8gb. I have installed local emoncms which also require some space for the db if you want it to run for years and not fill up the sd. My smallest size sd that i have is 15gb. @plonkster if you have a better way, please inform me. I know you can expand the file system. I am no Linux pro. Is there a way to shrink the file system ? Then i can make a smaller image for people to download and they can expand it themself
January 10, 20179 yr 5 minutes ago, Manie said: @plonkster if you have a better way, please inform me. I know you can expand the file system. I am no Linux pro. Is there a way to shrink the file system ? Then i can make a smaller image for people to download and they can expand it themself My criticism was somewhat tongue in cheek... no worries. I know roughly what the basic image size is. You need at least a 4GB card to fit the basic raspbian image, and when a manufacturer cheats (by calling it a 4GB card when it is only 3.9ish) that doesn't work either. There are two tricks to make it smaller. The first is to fill up all the free space on the image with zeroes before compressing, and there are also two ways to do this. You can either do this on a running image, eg: dd if=/dev/zero bs=1M of=spacer rm spacer The dd command will fail when it runs out of disk space. You then delete the file and it leaves the underlying blocks all zeroed. This makes all the free space on the image compress to almost nothing. You can also do that on the image itself. For that you need to do a bit of loopback work, using losetup. First, you need to print the partition table of your image, eg using a venus image: $ fdisk -l venus-image-raspberrypi2.rpi-sdimg Disk venus-image-raspberrypi2.rpi-sdimg: 748 MiB, 784334848 bytes, 1531904 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disklabel type: dos Disk identifier: 0xc60460e6 Device Boot Start End Sectors Size Id Type venus-image-raspberrypi2.rpi-sdimg1 * 8192 90111 81920 40M c W95 FAT32 (LBA) venus-image-raspberrypi2.rpi-sdimg2 90112 655359 565248 276M 83 Linux venus-image-raspberrypi2.rpi-sdimg3 655360 1220607 565248 276M 83 Linux venus-image-raspberrypi2.rpi-sdimg4 1220608 1531903 311296 152M 83 Linux Now the partition I want to zero is the second one, so I need to set up a loopback with an offset at 90112 in this example: $ sudo losetup /dev/loop0 venus-image-raspberrypi2.rpi-sdimg -o $((90112*512)) $ sudo file -s /dev/loop0 /dev/loop0: Linux rev 1.0 ext4 filesystem data, UUID=92e8a174-99af-405c-9944-23cf2e0d30b9 (extents) (huge files) Now you can mount loop0, and write files to it, eg: $ sudo mount /dev/loop0 /mnt $ sudo dd if=/dev/zero bs=1M of=/mnt/spacer $ sudo rm /mnt/spacer $ sudo umount /mnt At this point I prefer to give the file system a forced fsck so that it leaves the factory with a mount count of zero: $ sudo fsck -f /dev/loop0 Then you need only delete the loop device when done: $ sudo losetup -d /dev/loop0 The second way to make it smaller involves actually making the partitions smaller. That involves first using resize2fs (to make the file system smaller) and then fdisk (to make the partition smaller too), taking care not to make the partition smaller than the filesystem living on it. I'd rather leave that out unless there is someone who really really needs this to be smaller :-)
January 10, 20179 yr You can also (on the running image) remove some software that is not required. You'd first run dpkg -l to get such a list, eg: pi@raspberrypi ~ $ dpkg -l Desired=Unknown/Install/Remove/Purge/Hold | Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend |/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad) ||/ Name Version Architecture Description +++-=====================================-=========================================-============-=============================================================================== ii acl 2.2.52-2 armhf Access control list utilities ii adduser 3.113+nmu3 all add and remove users and groups ii aggregate 1.6-7 armhf ipv4 cidr prefix aggregator rc alsa-base 1.0.27+1 all dummy package to ease purging of obsolete conffiles ii apt 1.0.9.8.3 armhf commandline package manager . . . etc And then you can uninstall the things that are not really needed, eg if you're not doing any python development you can drop python2.7-dev: apt-get --purge remove python2.7-dev And so on and so forth. I recall that the default raspbian image is rather fully featured :-) Combining such a cleanup with zeroed blocks should bring it down somewhat.
January 10, 20179 yr I have added the pip4048 , as it is slightly different form the Axperts .A guy in Belgium is testing it and will report back.
January 10, 20179 yr 9 hours ago, plonkster said: Perhaps the sdcard automounts. That always happens on my system. You have to unmount it first and then attempt to write to it. That error is fairly generic, it's what happens when a syscall returns ENOTSUP. So it could be anything, but you could try and trace it with strace :-) You did: sudo dd if=imagefile bs=1M of=/dev/mmcblk0 right? yes (more or less) Pedro-MacBook:~ Pgutti$ sudo dd if=/Users/Pgutti/Desktop/ICC-PI.img of=/dev/disk4 bs=1024k dd: /dev/disk4: Resource busy Pedro-MacBook:~ Pgutti$ sudo dd if=/Users/Pgutti/Desktop/ICC-PI.img of=/dev/disk4s1 bs=1024k dd: /dev/disk4s1: Resource busy I was wable to format the SD with SD Formatter tool and i can copy files on it. But "dd" cannot write the img. Edited January 10, 20179 yr by PeterGutti attached screen shot
January 10, 20179 yr ..what is "strace" ? 10 hours ago, Manie said: Sorry i did not mention , the card must be at least 15gb That's ok: SD Card is 32GB.
January 10, 20179 yr 9 hours ago, plonkster said: Perhaps the sdcard automounts. That always happens on my system. You have to unmount it first and then attempt to write to it. That error is fairly generic, it's what happens when a syscall returns ENOTSUP. So it could be anything, but you could try and trace it with strace :-) You did: sudo dd if=imagefile bs=1M of=/dev/mmcblk0 right? what is "strace"?
January 10, 20179 yr 10 minutes ago, PeterGutti said: what is "strace"? It's a debugging tool that tracks all system calls and prints it out on the screen. You just prefix whatever command you're running with strace, eg "sudo strace dd....". The idea is that you can then see whatever system call is returning the ENOTSUP error. Not that that would help much to the average guy... but the result could be of use to a developer. I don't know if and how that works on the mac though.
January 10, 20179 yr plonkster, in mac os strace does not exist. But there is a consol tool: 10/1/17 22:49:51,719 sudo[5901]: Pgutti : TTY=ttys000 ; PWD=/Users/Pgutti ; USER=root ; COMMAND=/bin/dd if=/Users/Pgutti/Desktop/ICC-PI.img of=/dev/disk4 bs=1024k but there it shows only the command.. no reason for failing.
January 11, 20179 yr 10 hours ago, PeterGutti said: plonkster, in mac os strace does not exist. Indeed, I found that out this morning. There is something called dtruss that looks like it does the same thing, but it needs elevated privileges (on Linux it doesn't). Anyway... probably a wild goose chase anyway.
January 11, 20179 yr 20 hours ago, Manie said: I have added the pip4048 , as it is slightly different form the Axperts .A guy in Belgium is testing it and will report back. hi guys (girls), iam the guy from Belgium here in europe its almost all PIP4048 (ms / msd) versions, so the Axpert 5kv parallel was working , but not that good, Manie did here a very good job on the coding yesterday!! i would give you a kiss What is not working : inverter_frequency = not updating in emon (that you already know ) (not even with the last version you send to me) i connected my solarpanels today (i have dual MPPT), and i see only PV1 in the program and the "total" in emoncms i did send you an email today with the return code, i hope it helps. what would be great to see in the program and in emoncms, PV1 Amps, PV1 Volts, PV2 Amps , PV2 Volts, and PV1 + PV2 together (volts and amps) i have one (or some more ) word for ya Manie, keep it up and thanks for helping !!! talk to you later african brother grtz kristof
January 11, 20179 yr Thanks for the feedback yea , the PIP 4048 is still in testing face, I did the coding yesterday. I will still see how we can achieve this as i don't have a protocol doc specifically for yours, nor do i have the inverter.
January 11, 20179 yr Just now, Manie said: Thanks for the feedback yea , the PIP 4048 is still in testing face, I did the coding yesterday. I will still see how we can achieve this as i don't have a protocol doc specifically for yours, nor do i have the inverter. thats why iam here to test-run it
January 11, 20179 yr Please indicate to me in the return string what values should be the pv1 and pv2 etc. It all depends what the inverter return. If you have more than one Axpert in serial the temperature does not return from the inverter. If you can't see it in the return string then I can't help as the inverter return the info. There is lot that some Inverters does not return, that we would love to have. Like the inverter temp per inverter , not only one inverter.
January 11, 20179 yr 15 minutes ago, Manie said: Please indicate to me in the return string what values should be the pv1 and pv2 etc. It all depends what the inverter return. If you have more than one Axpert in serial the temperature does not return from the inverter. If you can't see it in the return string then I can't help as the inverter return the info. There is lot that some Inverters does not return, that we would love to have. Like the inverter temp per inverter , not only one inverter. the temp is working fine now so you dont need to worry about that anymore that was the return string i emailed you (227.4 49.9 229.8 49.9 0712 0506 014 397 49.70 000 055 0044 0001 104.5 49.70 00009 01010110 00 00 00054 010N{ on my inverter it was (when i was looking after i toke the screenshot) pv1 =54w and 107v so the value close to it in the return is = 055 0044 0001 104.5 49.70 could this be (at that time ) string 2 : 49.70 = 49.70v 0001 = 1watts string 1 : 104.5 = 104.5 0044 = 44w EDIT: pv1 is reading good, so you could know the code for that.. only pv2 is not reading out...
January 11, 20179 yr 5 minutes ago, kg666 said: on my inverter it was pv1 =54w and 107v so the value close to it in the return is = 055 0044 0001 104.5 49.70 could this be string 2 : 055 = 55w 0044 = 44watts nope it is battery capacity. Look the protocol doc and compare the values. Yours does not display the pv 1 and 2 separated
January 11, 20179 yr 1 4 minutes ago, Manie said: nope it is battery capacity. Look the protocol doc and compare the values. Yours does not display the pv 1 and 2 separated euhm not sure about that...my inverter can show pv1 and pv2 separated and total on the display when i looked in your program..that was the exact value of my PV1 on the inverter... not the combined voltage and watts the protocal doc you have is for the MS version, and that is indeed a single MPPT solar input, i have dual remember
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.