[SOLVED] Test conversion to systemd - Help with iptables status

With all the discussion around systemd and with a move for systemd to [core] soon, I thought I would convert an arch laptop from initscripts to systemd and see how it worked.
I followed the advice on the arch wiki for systemd, having already moved the systemd compatible parameters out of rc.conf a while ago. After installing systemd and systemd-arch-units, and adding init=/bin/systemd to the boot kernel line the system booted up nicely with systemd - graphical login failed as the relevant unit is not enabled at that stage - so hitting return to get a login prompt and logging as root it was quick to enable graphical.target and kdm.service, and rebooting then gave me a normal kdm login screen and then a normal login to kde.
All fine at this point. (By the way total login time was just over 30 seconds for both initscripts and systemd) - then it was necessary to check which services were running - most were running by default under systemd including postfix, dovecot, and sound worked fine with a quick check on a browser for a video feed. Networking was working fine too.
However iptables which was previously working fine under initscripts was not running - I checked that the firewall rules file was intact which is was.  Executing systemctl enable iptables.service appeared to be accepted but after executing systemctl start iptables.service and checking status again it was "exited" - the commands listed under the "start" command seemed normal and no error message appeared in the /var/log/messages file about iptables.
Can anyone suggest what is needed to get iptables working under systemd (yes I have read the wiki entries but nothing seemed to be relevant to fixing that particular issue).
Thanks in advance.
Last edited by mcloaked (2012-08-30 09:01:25)

cfr wrote:
I think what you're seeing is normal. Does it give a 0 exit status?
The service is of type "oneshot". I *think* that means that it is expected to exit.
As I understand it, iptables is a bit like ipfw. It's in the kernel and the userspace tools just provide an interface to that. So there's no daemon to run.
You can tell if it has worked correctly by
iptables --list
to check that your rules have been loaded correctly.
Note: I hope somebody will quickly correct me if I'm wrong about this because I've been assuming this is the expected behaviour on my own machine!
Perfectly correct - I didn't think to check iptables --list but yes all the rules are there as they should be.
It would be nice to have this documented in the wiki under systemd to help anyone else coming across the same issue.
I will mark this as solved in the topic title - thank you for replying - it was very helpful (and stopped me continuing the search for a solution by googling!)

Similar Messages

  • [SOLVED] python-efl: command 'gcc' failed with exit status 1

    Hi everyone! I'm running an x86_64 system with Enlightenment. I'm trying to install econnman from source to have a simpler life trying to connect to my University's wifi network. However, it misses the python-efl dependency. I downloaded it from the AUR, but I get the error message "error: command 'gcc' failed with exit status 1"  when running 'python setup.py build'. This is the complete terminology output:
    [dude@susan python-efl-1.8.0]$ sudo python setup.py build
    Checking for Eina: OK, found 1.11.2
    Checking for Eo: OK, found 1.11.2
    Checking for Evas: OK, found 1.11.2
    Checking for Ecore: OK, found 1.11.2
    Checking for EcoreFile: OK, found 1.11.2
    Checking for Edje: OK, found 1.11.2
    Checking for Emotion: OK, found 1.11.2
    Checking for DBus: OK, found 1.2.0
    Checking for Elementary: OK, found 1.11.2
    running build
    running build_py
    creating build/lib.linux-x86_64-3.4/e_dbus
    copying compat/e_dbus/__init__.py -> build/lib.linux-x86_64-3.4/e_dbus
    running build_ext
    building 'eo' extension
    gcc -pthread -Wno-unused-result -Werror=declaration-after-statement -DDYNAMIC_ANNOTATIONS_ENABLED=1 -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -march=x86-64 -mtune=generic -O2 -pipe -fstack-protector-strong --param=ssp-buffer-size=4 -fPIC -DEFL_BETA_API_SUPPORT -Iinclude/ -I/usr/include/python3.4m -c efl/eo/efl.eo.c -o build/temp.linux-x86_64-3.4/efl/eo/efl.eo.o -I/usr/include/efl-1 -I/usr/include/eo-1 -I/usr/include/efl-1 -I/usr/include/eina-1 -I/usr/include/eina-1/eina
    efl/eo/efl.eo.c: In function ‘__pyx_pf_3efl_2eo_4event_global_freeze_get’:
    efl/eo/efl.eo.c:1127:3: warning: implicit declaration of function ‘eo_event_global_freeze_get’ [-Wimplicit-function-declaration]
    eo_do(((const Eo *)eo_base_class_get()), eo_event_global_freeze_get((&__pyx_v_fcount)));
    ^
    efl/eo/efl.eo.c: In function ‘__pyx_f_3efl_2eo_object_from_instance’:
    efl/eo/efl.eo.c:1445:3: warning: implicit declaration of function ‘eo_base_data_get’ [-Wimplicit-function-declaration]
    eo_do(__pyx_v_obj, eo_base_data_get(__pyx_k_5, (&__pyx_v_data)));
    ^
    efl/eo/efl.eo.c: In function ‘__pyx_f_3efl_2eo__eo_event_del_cb’:
    efl/eo/efl.eo.c:2050:3: warning: implicit declaration of function ‘eo_base_data_del’ [-Wimplicit-function-declaration]
    eo_do(__pyx_v_self->obj, eo_base_data_del(__pyx_k_5));
    ^
    In file included from efl/eo/efl.eo.c:317:0:
    efl/eo/efl.eo.c: In function ‘__pyx_pf_3efl_2eo_2Eo_4__repr__’:
    efl/eo/efl.eo.c:2286:30: error: too many arguments to function ‘eo_parent_get’
    eo_do(__pyx_v_self->obj, eo_parent_get((&__pyx_v_parent)));
    ^
    /usr/include/eo-1/Eo.h:588:8: note: in definition of macro ‘_eo_do_common’
    __VA_ARGS__; \
    ^
    efl/eo/efl.eo.c:2286:5: note: in expansion of macro ‘eo_do’
    eo_do(__pyx_v_self->obj, eo_parent_get((&__pyx_v_parent)));
    ^
    In file included from /usr/include/eo-1/Eo.h:1112:0,
    from efl/eo/efl.eo.c:317:
    /usr/include/eo-1/eo_base.eo.h:37:12: note: declared here
    EOAPI Eo * eo_parent_get(void);
    ^
    efl/eo/efl.eo.c: In function ‘__pyx_f_3efl_2eo_2Eo__set_obj’:
    efl/eo/efl.eo.c:2496:3: warning: implicit declaration of function ‘eo_base_data_set’ [-Wimplicit-function-declaration]
    eo_do(__pyx_v_self->obj, eo_base_data_set(__pyx_k_5, ((void *)__pyx_v_self), NULL));
    ^
    In file included from efl/eo/efl.eo.c:317:0:
    efl/eo/efl.eo.c: In function ‘__pyx_pf_3efl_2eo_2Eo_14parent_get’:
    efl/eo/efl.eo.c:2963:28: error: too many arguments to function ‘eo_parent_get’
    eo_do(__pyx_v_self->obj, eo_parent_get((&__pyx_v_parent)));
    ^
    /usr/include/eo-1/Eo.h:588:8: note: in definition of macro ‘_eo_do_common’
    __VA_ARGS__; \
    ^
    efl/eo/efl.eo.c:2963:3: note: in expansion of macro ‘eo_do’
    eo_do(__pyx_v_self->obj, eo_parent_get((&__pyx_v_parent)));
    ^
    In file included from /usr/include/eo-1/Eo.h:1112:0,
    from efl/eo/efl.eo.c:317:
    /usr/include/eo-1/eo_base.eo.h:37:12: note: declared here
    EOAPI Eo * eo_parent_get(void);
    ^
    efl/eo/efl.eo.c: In function ‘__pyx_pf_3efl_2eo_2Eo_20event_freeze_get’:
    efl/eo/efl.eo.c:3104:3: warning: implicit declaration of function ‘eo_event_freeze_get’ [-Wimplicit-function-declaration]
    eo_do(__pyx_v_self->obj, eo_event_freeze_get((&__pyx_v_fcount)));
    ^
    error: command 'gcc' failed with exit status 1
    Can anyone help me? Please tell me every output you need. Thanks and a good day to you, sir!
    Last edited by HisDudeness (2014-10-02 08:47:15)

    You know what? You're right!
    I don't know how I exactly got that version... I got back to the AUR and downloaded the new one. The strange thing is, while that old version's tar.gz conained a lot of files like I'd expect from a source archive, the current version only contained the files .AURINFO and PKGBUILD. Guess that's the way it should be, I'm new to all of this so I was initially surprised. Obviously, running makepkg here worked like a charm and downloaded everything that was needed for it to install. Then, econnman installed just fine too. What I can't explain to myself is why 1.8.0's tar.gz contained all those files.
    In the end, I couldn't use nor connman cli nor econnman to connect to my Campus' WiFi but, after a lot of trouble, I got it to work with dhclient and wpa_supplicant like it's suggested. This hasn't been useless anyway, I saw how to install from the AUR for the first time.
    I'll mark the post as solved, but I got one last question: provided I don't need econnman anymore, do you suggest me to keep python-efl or to remove it? Is it needed for something else I'll certainly run into in the future?

  • Systemd - help with creating a service file

    Hi, I am trying to make a service file to execute "devmon --no-mount --exec-on-audio abcde". Basicly I want abcde to run when an audio cd is inserted into a headless computer. I have tried some udev rules but had no luck with that.
    I then found udevil and devmon, and it seems to do exectly what I want.
    So Ive made a systemd service file and put it in to /usr/lib/systemd/system
    here is it:
    /usr/lib/systemd/system/abcde-auto.service
    [Unit]
    Description=Virtual Distributed Ethernet
    [Service]
    Type=simple
    ExecStart=/usr/bin/devmon --no-mount --exec-on-audio abcde
    [Install]
    WantedBy=multi-user.target
    I have started this with
    systemctl start abcde-auto.service
    and I see that a devmon process has started with the same arguments I gave in the service file, so all is good.
    But now, I am not too sure about the "Type" in the service file, devmon starts abcde, but:
    systemctl stop abcde-auto.service
    does not stop abcde, only devmon.
    Is there a way to setup the service file so that abcde will end as well as devmon?
    *Edit
    Ive read this http://www.freedesktop.org/software/sys … rvice.html but Im still not too sure. Dont really know what most of it means.
    Last edited by jrussell (2012-10-20 18:09:23)

    I use type 'oneshot' and add an ExecStop= to start and stop scripts.
    Like here is the unit file I use to start and stop arno-iptables-firewall
    [Unit]
    Description=Arno iptables firewall
    [Service]
    Type=oneshot
    RemainAfterExit=yes
    ExecStart=/usr/sbin/arno-iptables-firewall start
    ExecStop=/usr/sbin/arno-iptables-firewall stop
    [Install]
    WantedBy=multi-user.target

  • Working in test environment. Need help with creating a notification subscription

    I am pulling my hair out here. Whatever I have left. I am experimenting in my test environment. I am trying to create a notification subscription. Under criteria, when I select Support Group, I only have a choice of 3. Since my test is a copy of Prod, I
    am confused here because in Prod, I have about 20 Support Groups to choose from. I found a List under Library --> Lists called Incident Tier Queue and the 3 Support Groups I see are in this list. Thing is, I can't edit the list. I get some kind of validation
    error. What I would like to know is when you are creating a Subscription and you select Support Group in the Criteria, where is it looking or what List if any is it looking at to get that information and how would I change it?

    if OOB when creating an email subscription the Support Group check box above points to the List Incident Tier Queue, how does one change that to point to the Service Request Support Group? There has to be a change somewhere else that allows it to point to
    the customized list.
    You can't change that. The Support Group property defined on the Incident class is an enumeration property that uses the IncidentTierQueue enumeration (list). This property's proper name is "TierQueue".
    The Support Group property defined on the Service Request class is an enumeration property that uses the ServiceRequestSupportGroupEnum enumeration (list). This property's proper name is "SupportGroup".
    Those two properties are unrelated.
    You can't "point" an existing property to a different list. You can only modify the values in the list by adding values in an unsealed management pack.
    If, in your production environment, you have a bunch of options in your Incident's "Support Group" property beyond Tier 1, Tier 2, and Tier 3, then somewhere in your production environment is an MP that contains those new properties. That MP happens
    to be the "ServiceManager.IncidentManagement.Configuration" MP. The friendly name, in the management packs list, is "Service Manager Incident Management Configuration Library".
    Now, as you say, if there are a bunch of list values in the Incident support group in your production environment, then you need to import that MP into your test environment. Then, when managing Incident criteria, you'll be able to choose from all of those
    options.
    Furthermore..let's say that your support groups are only defined in the ServiceRequestSupportGroupEnum..then you can't use them in your Incident criteria. Neither in your test nor your production environments. The IncidentTierQueue enum has to be updated
    with any options you want to use during Incident subscription creation.
    Lastly (if the above is true)..are you sure you want to create an Incident subscription and not a Service Request subscription?

  • [SOLVED]Would be grateful for help with im notification shell script

    Hi all,
    Playing with finch (a console im client), and haven't found any notification plugins for it that work the way I want.
    What I have been doing is using tail to watch the logs. As the logs are in html, I use sed to strip the html tags:
    tail -n 1 -q -f ~/.purple/logs/bonjour/phinch/*/* | sed -e 's#<[^>]*>##g'
    I would like to use notify-send or dzen2 to create popup windows when the log file changes, but I'm not sure how (I am not an experienced programmer).
    This, for example, does not work:
    tail -n 1 -q -f ~/.purple/logs/bonjour/phinch/*/* | sed -e 's#<[^>]*>##g' | while read line ; do notify-send "${line}" ; done
    Removing sed from the stream above allows notify-send to pop up notifications. How can I have the same happen with sed in there?
    Last edited by hellomynameisphil (2011-10-22 17:49:34)

    hellomynameisphil wrote:
    tail -n 1 -q -f ~/.purple/logs/bonjour/phinch/*/* | sed -e 's#<[^>]*>##g' | while read line ; do notify-send "${line}" ; done
    Removing sed from the stream above allows notify-send to pop up notifications. How can I have the same happen with sed in there?
    The problem here is that sed is buffering its output to improve efficiency in pipes. Try with:
    tail -n 1 -q -f ~/.purple/logs/bonjour/phinch/*/* | sed -e 's#<[^>]*>##g' -u | while read line ; do notify-send "${line}" ; done
    And should work as well
    Regards

  • New JDWP Implementation - help with suspend status

    Hi All,
    My first post here.
    I am presently implementing JDWP into a small VM - no JVMTI underneath. I am having trouble understanding the interaction between debugger (eclipse 3.4) and debuggee (the VM).
    The VM starts a single thread and suspends all while it waits for a incoming connection. A connection comes in from eclipse and debugger/debuggee make their way through all commands/replies up the the VirtualMachine_Resume command to get things started again.
    During all this, the debugger requests a event set CLASS_PREPARE on a class with a classmatch filter and a SuspendPolicy of THREAD. Fair enough.
    So ... the VM sends the CLASS_PREPARE command to the debugger and then suspends the single thread. As there is only one thread, the VM has to basically spin waiting for something from the debugger to resume the thread (I guess ??).
    BUT .. then the debugger nevers sends anything.
    Is my understanding correct? Should the debugger respond to get stuff going again? (yes - "I'm doing something wrong", no = "I don't understand, please explain ...")
    All help appreciated.
    Greg.

    Self-reply .. for the record and as a note to other (unfortunate) persons who are implementing JDWP from the spec.
    My issue was with what happened when the VM started. I was performing the startup VMStart interaction before the main class had been loaded. It is wise to start the VM/debugger interaction after the main class has been loaded - escpecially if your testing involves putting a breakpoint in the main class. :-/
    G.

  • Help with combined status query

    I have another puzzler...
    I have the following:
    select * from stat;
    ID_NO STAT                 BEGDT       ENDDT
        1 Active               01-JAN-2012 31-DEC-4712
    select * from loa;
    ID_NO STAT                 BEGDT       ENDDT
        1 Paid                 01-FEB-2012 14-MAR-2012
        1 UnPaid               17-MAY-2012 29-MAY-2012I want the following:
    ID_NO  STAT             BEGDT             ENDDT
    1      Active           01-JAN-2012       31-JAN-2012
    1      Paid             01-FEB-2012       14-MAR-2012
    1      Active           15-MAR-2012       16-MAY-2012
    1      UnPaid           17-MAY-2012       29-MAY-2012
    1      Active           30-MAY-2012       31-DEC-4712Sample data:
    Version:
    Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - 64bit Production
    PL/SQL Release 11.2.0.1.0 - Production
    CORE 11.2.0.1.0 Production
    TNS for Linux: Version 11.2.0.1.0 - Production
    NLSRTL Version 11.2.0.1.0 - Production
    create table stat( id_no  number(3)
                   , stat   varchar2(20)
                   , begdt  date
                   , enddt  date
    create table loa( id_no  number(3)
                   , stat   varchar2(20)
                   , begdt  date
                   , enddt  date
    insert into stat values(1,   'Active',  to_date('01-JAN-2012', 'DD-MON-YYYY'), to_date('31-DEC-4712', 'DD-MON-YYYY'));
    insert into loa  values(1,   'Paid',    to_date('01-FEB-2012', 'DD-MON-YYYY'), to_date('14-MAR-2012', 'DD-MON-YYYY'));
    insert into loa  values(1,   'UnPaid',  to_date('17-MAY-2012', 'DD-MON-YYYY'), to_date('29-MAY-2012', 'DD-MON-YYYY'));I'm drawing a complete blank...
    Once again, any help would be appreciated.

    Hi,
    Welcome to the forum!
    Here's one way:
    WITH     all_events     AS
         SELECT     id_no, stat, dt, loa_lvl
         ,     stat          AS base_stat
         FROM     stat
         UNPIVOT     (     dt
              FOR      loa_lvl IN ( begdt     AS 0
                           , enddt     AS 9
        UNION ALL
         SELECT     id_no, stat, dt, loa_lvl
         ,     NULL          AS base_stat
         FROM     loa
         UNPIVOT     (     dt
              FOR      loa_lvl IN ( begdt     AS 1
                           , enddt     AS -1
    ,     got_analytics     AS
         SELECT     id_no, dt, stat
         ,      LAST_VALUE (base_stat IGNORE NULLS)
                         OVER ( PARTITION BY  id_no
                                       ORDER BY          dt
                           ) AS last_base_stat
         ,     SUM (loa_lvl) OVER ( PARTITION BY  id_no
                                       ORDER BY          dt
                           ) AS loa_sum
         ,     LEAD (dt)     OVER ( PARTITION BY  id_no
                                       ORDER BY          dt
                           ) AS next_dt
         FROM    all_events
    SELECT       id_no
    ,       CASE
               WHEN  loa_sum  > 0 THEN stat
                                          ELSE last_base_stat
           END          AS stat
    ,       CASE
              WHEN  loa_sum = 0
              AND   dt > MIN (dt) OVER (PARTITION BY id_no)
                                    THEN  dt + 1
                                       ELSE  dt
           END          AS begdt
    ,       CASE
              WHEN  loa_sum = 0
              AND    dt < MAX (dt) OVER (PARTITION BY id_no)
                                        THEN  next_dt - 1
                                       ELSE  next_dt
           END          AS enddt
    FROM       got_analytics
    WHERE       next_dt     IS NOT NULL
    ORDER BY  id_no
    ,            dt
    ;This assumes that each id_no is a world unto itself, uninfluenced by other id_nos
    Can rows in loa with the same id_no overlap? That is, can the begdt or enddt of one row be between the begdt and enddt of another row in loa with the same id_no?
    Can you have 2 (or more) rows in stat with the same id_no? Can they ovelap?
    If you answered "Yes" to any of the above, include examples in your sample data and results. The query may need to be adjusted.
    I wish we could mark questions as being "Correct"; you'd have at least 10 points for your first thread. Thanks for including the CREATE TABLE and INSERT statemnts, the very clear results, and your version information, all neatly formatted. Keep up the good job!
    Edited by: Frank Kulash on Mar 23, 2012 2:53 PM

  • Help with WBS status management

    Hi,
    I am trying to create a new user status for a WBS in OKO2 which will forbid material postings through requisitions and orders and also forbid sales orders to be created. These are standard business transactions and I can see are easily available to forbid when creating the new status.
    It works when trying to create a new material document or sales order.
    However, can anyone help as I am also trying to ensure that if a material posting has been made already to the WBS or sales order is assigned and then someone tries to change the user status to this new user status, then the validation of the project on save will create an error message an advise that the status can not be set due to an existing material document or sales order.
    I would like to do this without having to create an additional validation user exit.
    Can anyone advise if this is possible?
    Thanks

    For postings that are already made this can't be achieved. As suggested by Abdul you may have to use Validations or any other work around to achieve this.

  • Sales doc items with the status CLSD ARE being selected for month-end

    We are working to improve month-end processing time for Sales Orders  We use SAPKKA12 to set CLSD status on line items
    SAP Help states: be sure that the status CLSD (closed) is set for all sales document items that you do not want to include in the period-end closing process. The sales document items with the status CLSD are then not selected for processing in the period-end closing process. Use report program SAPKKA12 to set the status.
    This use to work (in 4.0). I have been on other projects and just brought back to this after several years (now in 5.0).  Testing today shows that orders with CLSD status are being selected for month-end process.
    Any ideas why this is so?  System is selceting over 300,000 orders when well over half of them are set to CLSD.  I do get a message that the order is not relevant to processing, but selecting the order and logging the message take time. and the logs are VERY long.
    thanks

    Does any one else have this issue  that Sales documents with status CLSD are selected by RA (KKAK)?
    In just one of our jobs, i get this Basic List
    Perform actual results analysis      31273      
    Not relevant                                                 103894
    Error                                                        2306      
    Objects Selected                                    137,473
    75% of the orders are CLSD.  and the system is still counting them. (we do not out put information messages)
    I do understand that the orders are not analyzed,  but all documentaion states that CLSD status is not selected, and in our 5.0 system, they are selected.  In our 4.0 system, they were not selected.
    Does this truely work for anyone in a 5.0 system?
    Thanks
    Althea

  • Help with systemd problems/reply to closed thread

    @Everyone; as the previopus topic was closed, I am opening a new one here, partly to reply to some things that was said to me in the previous thread/topic (and I do not like to leave people hanging), but mainly to try and get some help with my systemd config (that I was not able to find help with in the wiki or manpages).
    @czubek; eww, that sounds rather similiar to some other operating systems I've heard of (not just, but obviously including, windows)... For me, subjectively, that is not very attractive, as being able to have a single application do a single thing, and then mixing and matching as I personally see fit is a great thing.
    @tomegun: I can indeed see your point, but if I made direct quotes on all the things I reply too, my posts would reach biblical proportions.
    Mmm-hmm, I do indeed know of "the Arch Way"; however, the system that you expose can be of varying complexity as well, and I found that the system that was presented to me by way of Sysvinit/initscripts was a way simpler system than that presented to me by systemd.
    Indeed, and I do like that; I enjoy knowing how my system works and learning more about it; however, the truth is that reading manpages and the wiki isn't always enough. And I use the numbers instead of letters in certain places as a jest, please, do not get offended by that; if you have to, get offended by my opinions.
    Ah, I have now scanned the wiki, the manpages, and have already faced problems. While part of me is saddened by this fact (as I was hoping for a smooth transition to systemd), part of me is glad that I can at least prove that everything isn't quite as simple as some people would make it out to be. More on this later in this post, as I do need help with the problems I have faced.
    And I wouldn't say that the new and old way are equally trivial, as I immediatley faced problems with "the new way". More on this later.
    Ok, I find that if the debate is polarized, then there is nothing wrong with describing it as such. I spell what I see.
    Oh my... Ignored as a matter of principle? I would say that that is an excellent way to stagnate; by that logic, moviemakers, gamemakers etcetera should ignore any and all criticism they recieve, as the critics are not contributing.... Wouldn't that lead to a complete halt in development? Isn't criticism (so long as it is constructive, obviously) a key component of (media/software) evolution?
    Ah, I see, then I  misunderstood the current state of the install media
    "Some adjustements"... Well, I guess we'll have to see how that goes, then. Hopefully it'll be smooth.
    @Everyone; Awebb possesses stunning accuracy as far as his observations go.
    @zb3; My point exactly.
    @tomegun; I believe you to be wrong; complaining (but not whining) is the first step towards improvement, as I think that I and zb3 has shown.
    Further, your last point (where you talk about "deep understanding of Y" and whatnot) is basically saying that only devs and programmers can contribute, so zb3's observation would be correct.
    Also, making suggestions is part of constructive criticism; you can do that even if you can't code, e.g. by saying "X works badly because it makes Y crash; couldn't we replace X with something akin to Z?". This is a valid point even if you have no intricate understanding of X, Y or Z, and if it is impossible to replace X for one reason or the other, it's not hard to reply with "Replacing X is impossible because of reason Q", and then see if someone can solve reason Q etcetera.
    zb3's example is obviously helping; for one, it would allow people to know taht they can't boot with systemd in certain cases, or that installing/runing systemd can break their system. I would consider that rather important information.
    @Everyone; Now then, to see if I can (or rather, if I can get help to) shed some light on some problems I've been having.
    My transition to systemd went pretty smooth (besides the fact that I couldn't use 'systemctl' to set anything, and neither duckduckgo nor any posts in the forums could help me (the error I continously got was "Failed to issue method call: Launch helper exited with unknown return code 1") and that there are no manpages/tips on how to configure the files for "localtime" and "adjtime", which would've been good considering some syntax have changed) to about halfway through, to the point where I was supposed to start using systemd for my daemons.
    Now then, to the main problem; first, I tried running
    "systemctl enable syslog-ng.service"
    and got the following message;
    "ln -s '/usr/lib/systemd/system/syslog-ng.service' '/etc/systemd/system/syslog.service'
    ln -s '/usr/lib/systemd/system/syslog-ng.service' '/etc/systemd/system/multi-user.target.wants/syslog-ng.service'"
    Since it doesn't exclusively say that something went wrong (at least as far as I can understand), I pondered the message a bit, but moved on to try and start dbus by issuing
    "systemctl enable dbus.service",
    to which I got the reply
    "The unit files have no [Install] section. They are not meant to be enabled using systemctl."
    This I can only assume is an error message of some sort, especially seeing as it says "not meant to be enabled using systemctl", but the wiki rather explicitly states that that is what I am supposed to do.
    I dare not move on (or attempt to start other daemons) without further advice, as I fear I might break my system. Any ideas as to how to solve this would be greatly appreciated.

    tcmdvm wrote:
    The message;
    "ln -s '/usr/lib/systemd/system/syslog-ng.service' '/etc/systemd/system/syslog.service'
    ln -s '/usr/lib/systemd/system/syslog-ng.service' '/etc/systemd/system/multi-user.target.wants/syslog-ng.service'"
    indicates that running "systemctl enable syslog.service" is now enabled.
    If you try running sytemtctl enable <whatever>.service and get
    "The unit files have no [Install] section. They are not meant to be enabled using systemctl." means there is no <whatever>.service file available to enable.
    The dbus.service as far as I know is already enabled when installing systemd and doesn't require any action.
    Ah, right, systemd is chatty (I'm of the old skool; if nothing is/goes wrong, a program should keep quiet).
    Oh ok, then I should dare more on.
    ZekeSulastin wrote:To be a bit more precise, it means the service is there but it's only used as a dependency for something else; the [Install] section of the file is what tells systemctl where to link it and such.  Also, where exactly in the wiki were you told explicitly to 'systemctl enable dbus.service'?  I can easily point you to the sections in both installation guides that tell you how to set things like /etc/localtime et. al (also `man 7 archlinux`), but not the one that said to manually enable dbus.
    Ah, I see; and well, I was told right here, going from the top, to "Enable Daemons formerly listed in rc.conf [...] For a translation of the daemons from /etc/rc.conf to systemd services, see: List of Daemons and Services.", and, using the list, dbus is mentioned there, as you can see. Since I started from the top, I did not know that you didn't need to enable that service. Perhaps it could be added to the descriptions of the relevant services in the Wiki which ones are autostarted?
    ZekeSulastin wrote:Lastly, I don't think trying to continue conversations from a closed thread is a very good idea, let alone doing it in a giant undivided wall of text that makes it quite difficult to pick out much any single topic.
    Meh, I directly refer to the relevant people, they can easily see their own old posts in the previous thread, etcetera... I think te ones who have something to say will be able to find my replies to their previous statements (or so I hope). For now, as everyone can see, I quote stuff (but man might I make some huge posts if I ever post again in the future, if I quote what I reply to).
    fsckd wrote:This is not really a technical support subforum. Moving from Arch Discussion to Newbie Corner.
    Ah ok, thanks (or perhaps a cautious thanks? Moving it to "newbie corner" might, after all, be a subtle insult...).
    tomegun wrote:It is not really possible to follow your message as you don't quote what you are replying to. Moreover, as the thread was closed I think that means we should stop the discussion ;-)
    Hmm, I think it means cop-out for lack or proper arguments/responses... But ok, I'll let it slide, for now (mainly as I can't do much else).
    tomegun wrote:You are not meant to enable dbus.service as it is enabled unconditionally. To check the enabled/disabled state of your services try "systemctl list-unit-files". "static" means that you are not supposed to enabled/disable it.
    Ah ok, thank you. Now, having abandoned all hope, I'll start configuring again and see how it goes... May the void protect me. If I die, grieve not for me, but remember me in the fight against hard-to-configure applications.
    (edit/addon)
    Actually, what am I supposed to do with "hwclock"? the List of Daemons only says that I shouldn't run that in tandem with ntpd, but I'm not running ntpd, so... How do I get hwclock with systemd?
    Edited for additional question.
    Last edited by incassum (2012-11-07 08:44:42)

  • [SOLVED] Need a little help with ACPI handler.sh script.

    By using ACPI I'm trying to get my laptop to hibernate at critical power state. So far I've made a few test scripts to test how to do this, with one of these scripts I'm stuck. My scripting skills are not that good to solve it.
    This is the script:
    if [ `awk '{print $3}' /proc/acpi/battery/C1E9/state` == "ok" ]
    then
    logger "battery is ok"
    else
    logger "battery is not ok"
    fi
    If I run the script I get an error (while "battery is not ok" is logged):
    ┌─[~]
    └─> ./test2
    ./test2: line 1: [: too many arguments
    ┌─[~]
    └─>
    When I run the awk-command from the script, I get the following output:
    ┌─[~]
    └─> awk '{print $3}' /proc/acpi/battery/C1E9/state
    ok
    discharging
    2341
    1824
    10969
    ┌─[~]
    └─>
    I think the problem lies in the multiple output of the awk-command, but I'm stuck in solving it.
    Any help would be great.
    Last edited by NeoXP (2009-11-16 22:40:45)

    @ skanky
    Thanks for the suggestion, but you're right I'm not gonna change a working script. It is working exactly as expected.
    @ all
    When interested this is my current handler.sh
    #!/bin/sh
    # Default acpi script that takes an entry for all actions
    # NOTE: This is a 2.6-centric script. If you use 2.4.x, you'll have to
    # modify it to not use /sys
    minspeed=`cat /sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_min_freq`
    maxspeed=`cat /sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_max_freq`
    setspeed="/sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed"
    set $*
    case "$1" in
    button/power)
    echo "PowerButton pressed!">/dev/tty5
    case "$2" in
    PWRF) logger "PowerButton pressed: $2" ;;
    *) logger "ACPI action undefined: $2" ;;
    esac
    button/sleep)
    case "$2" in
    SLPB) echo -n mem >/sys/power/state ;;
    *) logger "ACPI action undefined: $2" ;;
    esac
    ac_adapter)
    case "$2" in
    AC)
    case "$4" in
    00000000)
    echo -n $minspeed >$setspeed
    #/etc/laptop-mode/laptop-mode start
    00000001)
    echo -n $maxspeed >$setspeed
    #/etc/laptop-mode/laptop-mode stop
    esac
    *) logger "ACPI action undefined: $2" ;;
    esac
    battery)
    if [ `awk '{print $2}' /proc/acpi/ac_adapter/C1E8/state` == "off-line" ];
    then
    logger "battery discharging"
    if (( `awk '{if (NR==5) {print $3}}' /proc/acpi/battery/C1E9/state` >= "250" ))
    then
    logger "battery ok"
    else
    logger "battery not ok"
    /usr/sbin/pm-hibernate;
    fi
    else logger "battery charging"
    fi
    button/lid)
    if [ `awk '{print $2}' /proc/acpi/button/lid/C1ED/state` == "closed" ];
    then
    logger "ACPI lid closed"
    /usr/sbin/pm-suspend
    fi
    logger "ACPI group/action undefined: $1 / $2"
    esac
    Still working on it though

  • Need Help with data type conversion

    Hello People,
    I am new to java, i need some help with data type conversion:
    I have variable(string) storing IP Address
    IPAddr="10.10.103.10"
    I have to call a library function which passes IP Address and does something and returns me a value.
    The problem I have is that external function call in this library excepts IP Address in form of a byte array.
    Here is the syntax for the function I am calling through my program
    int createDevice (byte[] ipAddress).
    now my problem is I don't know how to convert the string  IPAddr variable into a byte[] ipAddress to pass it through method.

    Class InetAddress has a method
    byte[]      getAddress() You can create an instance using the static method getByName() providing the IP address string as argument.

  • Help with "Testing Failed" Error message in iWeb09

    I bought the iLife 09 upgrade purely so I could publish to my own domain server. I have a MobileMe account and iWeb09 publishes fine there--no problems.
    My domain name is hosted on a Windows NT server. The admin is willing to help, but I was hoping someone could point me to documentation for this error:
    "Testing Failed.
    Unable to write to server. Verify your settings and try again."
    If I ftp to the site in a terminal window and do a put, I can write to the site. Obviously, I could at some point, because it servers the "dummy" page I put up, but I tested again in case my permissions had changed.
    I looked at the "Publishing to Windows-based FTP Server" but it does not seem to describe this error.
    BTW, I pay AppleCare for support, but when I tried to get support documentation from them, the tech said they "can't help" with this part of iWeb and hung up on me--the call lasted exactly 90 seconds.
    I am hoping someone out here will know where the errors are documented or will be able to help me figure it out.

    I waded through the posts for iWeb09 and came across a recommendation for Cyberduck (www.cyberduck.ch), so I tried a copy. Works like a champ. Unlike iWeb09, all I had to do was type in my server name, user name, and password to open the connection. Then I published my site to a folder in iWeb and uploaded the folder in Cyberduck. Cyberduck had no problem writing to the server.
    I still would love to know why iWeb can't write to my web server.

  • I need your help with solving a non-standard solutions.

    Hello!
    I need your help with solving a non-standard solutions.
    My name is Andrii and I represent the civil charity. Not so long ago, one man gave us the iPhone 4S (as a donation).
    But later - it became known that this iPhone can not work with SIM cards Ukrainian mobile network operators, as it is blocked by Verizon.
    Is there a way to unlock it with your help, and how it can be done?
    I hope for your understanding and assistance in this matter!
    Thank you.

    Contact a third party unlocking service.
    Purchase an unlocking code.

  • Lil' help with systemd user session units

    I'm using systemd 207. i would like to change my desktop-session on the fly. in this suituation form razor to openbox. these are my configs located in ~/.config/systemd/user/
    razor.config
    [Unit]
    Description=Razor Qt Desktop
    Before=mystuff.target
    After=xorg.target
    After=environment.target
    Requires=dbus.socket
    [Service]
    ExecStart=/usr/bin/startrazor
    #Restart=on-failure
    Environment=DISPLAY=:0
    KillMode=process
    [Install]
    WantedBy=de.target
    de.target
    [Unit]
    Description=Desktop Enviroments
    Wants=xorg.target
    Wants=mystuff.target
    Requires=dbus.socket
    AllowIsolate=true
    [Install]
    Alias=default.target
    openbox.service
    [Unit]
    Description=The Openbox Window Manager
    After=xorg.target
    Requires=dbus.socket
    [Service]
    ExecStart=/usr/bin/openbox
    Restart=on-failure
    [Install]
    WantedBy=openbox.target
    openbox.target
    [Unit]
    Description=Openbox
    Wants=xorg.target
    Wants=mystuff.target
    Requires=dbus.socket
    AllowIsolate=true
    mystuff.target
    [Unit]
    Description=Xinitrc Stuff
    After=environment.target
    Wants=razor.target
    Wants=environment.target
    [Install]
    Alias=default.target
    now...
    systemctl --user isolate razor.target
    ...doesen't close razor-qt desktop to make place for a new wm session
    what shall we do whith the system units....
    what shall we do whith the system units....
    what shall we do whith the system units....
    early in the morning
    Last edited by kriz (2013-09-25 12:25:43)

    I have this problem too.  I had not seen this thread you linked to above before... but in my one reboot test (which shoud be taken with a bucket of salt since my machine sometimes shuts down okay) it worked great.  It would seem that 65kid's workaround is effective.  I'm going to reboot again to see if this is consistent. I'll udpate if it hangs.
    Edit: So it didn't hang, but it looked like it was going to.  Often the session will seem to start to exit, but the root window will remain (the background).  It did just this, and then after a second, continued to the printing of the shutdown messages.  Success!
    Edit2: I guess I should note the actual fix so that others don't have to follow that link.  As putting the workaround in more than one place is probably optimal.  So you need to add "TimeoutStopSec=2" to the [Service] section of the [email protected].  This can be done either by the .include function of unit files, or by the drop-in replacement functionality via /etc/systemd/system/user-session@<youruser>.d/<added-config>.conf.  I would imagine that you could play around with the value of the timeout.  Systemd's time capabilities are pretty extensive, so I imagine that you could actually configure it down to the ms here.
    Last edited by WonderWoofy (2013-06-04 23:17:41)

Maybe you are looking for